EXPLANATORY SUPPLEMENT
TO THE ASTRONOMICAL ALMANAC
A Revision to the
EXPLANATORY SUPPLEMENT TO THE ASTRONOMICAL EPHEMERIS
and
THE AMERICAN EPHEMERIS AND NAUTICAL ALMANAC
Prepared by
THE NAUTICAL ALMANAC OFFICE, U.S. NAVAL OBSERVATORY
WITH CONTRIBUTIONS FROM
H.M. NAUTICAL ALMANAC OFFICE, ROYAL GREENWICH OBSERVATORY,
JET PROPULSION LABORATORY,
BUREAU DES LONGITUDES,
and
THE TIME SERVICE AND ASTROMETRY DEPARTMENTS, U.S. NAVAL OBSERVATORY
Edited by
P. Kenneth Seidelmann
UNIVERSITY SCIENCE BOOKS
Mill Valley, California
University Science Books
20 Edgehill Road
Mill Valley, CA 94941
Fax: (415)383-3167
Production manager: Mary Miller
Copy editor: Aidan Kelly and Margy Kuntz
Text and jacket designer: Robert Ishi
TgX formatter and illustrator: Ed Sznyter
Proofreader: Jan McDearmon
Printer and binder: The Maple- Vail Book Manufacturing Group
Copyright © 1992 by University Science Books
Reproduction or translation of any pan of this work beyond
that permitted by Section 107 or 108 of the 1976 United States
Copyright Act without the permission of the copyright owner
is unlawful. Requests for permission or further information
should be addressed to the Permissions Department,
University Science Books.
Library of Congress Catalog Number: 91-65331
ISBN 0-935702-68-7
Printed in the United States of America
10 987654321
Abbreviated CONTENTS
LIST OF FIGURES xvii
LIST OF TABLES xxi
FOREWORD xxv
PREFACE xxvii
1 INTRODUCTION TO POSITIONAL ASTRONOMY 1
2 TIME 39
3 CELESTIAL REFERENCE SYSTEMS 95
4 TERRESTRIAL COORDINATES
AND THE ROTATION OF THE EARTH 199
5 ORBITAL EPHEMERIDES
OF THE SUN. MOON, AND PLANETS 279
6 ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 325
7 PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND
SATELLITES 383
8 ECLIPSES OF THE SUN AND MOON 421
9 ASTRONOMICAL PHENOMENA 475
10 STARS AND STELLAR SYSTEMS 505
11 COMPUTATIONAL TECHNIQUES 541
12 CALENDARS 575
12.10 REFERENCES 606
13 HISTORICAL INFORMATION 609
14 RELATED PUBLICATIONS 667
15 REFERENCE DATA 693
GLOSSARY 721
INDEX 741
v
Contents
LIST OF FIGURES xvri
LIST OF TABLES xxi
FOREWORD xxv
PREFACE xxvii
1 / INTRODUCTION TO POSITIONAL ASTRONOMY 1
1.1 INTRODUCTION 1
1.11 Purpose 1
1 .2 TIMESCALES AND CALENDARS 2
1.21 Atomic Timescales 2
1 .22 Dynamical Time 3
1 .23 Rotational Timescales 3
1.24 Coordinated Universal Time (UTC) 6
1.25 The Enumeration of Dates 7
1.3 CELESTIAL AND TERRESTRIAL COORDINATES 8
1.31 Coordinate Systems and Frames 8
1.32 Celestial Coordinate Systems 11
1.33 Terrestrial Coordinate Systems 13
1 .34 The Rotation of the Earth 1 8
1 .35 The Connections between Terrestrial and Celestial Coordinates 20
1 .36 Effects of the Position and Motion of the Object and Observer 21
1 .4 ORBITAL MOTIONS 24
1 .41 Motion in Two-body Systems 24
1 .42 Types of Perturbations 28
1 .43 Perturbations by and on Extended Bodies 30
1.5 ASTRONOMICAL PHENOMENA 32
1.51 Rising, Setting, and Twilight 32
1.52 Meridian Transit 33
vil
viii
CONTENTS
1.53 Conjunction, Opposition, and Elongation 33
1 .54 Eclipses, Occultations, and Transits 34
1.55 Satellite Phenomena 35
1 .56 Physical Observations of the Sun, Moon, and Planets 36
1.6 REFERENCES 38
2 / TIME 39
2.1 INTRODUCTION 39
2.2 MEASURES OF TIME AND THEIR RELATIONSHIPS 40
2.21 Atomic Time (TAI) 40
2.22 Dynamical Time 41
2.23 Sidereal Time 48
2.24 Universal Time 50
2.25 The Ephemeris Meridian 54
2.26 Julian Date 55
2.27 Time Zones 56
2.3 PRACTICAL DETERMINATIONS OF TIME 58
2.31 Frequency Standards and Clock Performance 58
2.32 Measurement of Atomic Time 60
2.33 Earth Rotation Measurement 61
2.34 Dynamical Time Determinations 63
2.4 METHODS OF TIME TRANSFER 64
2.41 Radio Time Signals 65
2.42 Portable Clocks 65
2.43 LORAN-C 66
2.44 Television Comparison Techniques 67
2.45 Use of Satellites 68
2.46 Intercontinental Clock Synchronization by VLBI 69
2.47 Relativistic Effects in Time Transfer 70
2.5 HISTORICAL DEVELOPMENT OF TIMEKEEPING 73
2.51 Introduction 73
2.52 Apparent Solar Time, Mean Solar Time, and the Equation of Time 74
2.53 Rotation of the Earth 76
2.54 Universal Time 77
2.55 Ephemeris Time 79
2.56 History of Atomic Time 84
2.57 History of Coordinated Universal Time 85
2.58 History of Transmitted Time Signals 86
2.6 REFERENCES 88
3 / CELESTIAL REFERENCE SYSTEMS 95
3.1 CELESTIAL REFERENCE SYSTEMS 95
3.11 Fundamental Reference Systems 97
3.12 The Dynamical Reference System 98
3.13 The Conventional Celestial Reference System 98
3.2 BASIS OF REDUCTION OF CELESTIAL COORDINATES 99
3.21 Precession 99
CONTENTS
ix
3.22 Nutation 109
3.23 Space Motion 121
3.24 Parallax 123
3.25 Aberration 127
3.26 Gravitational light deflection 135
3.27 Polar Motion 139
3.28 Refraction 140
3.3 APPARENT AND TOPOCENTRIC PLACE ALGORITHMS 145
3.31 Apparent- Place Algorithm for Planets 1 45
3.32 Apparent-Place Algorithm for Stars 1 52
3.33 The Computer Implementation of Apparent-Place Algorithms 1 54
3.34 Apparent-Places-Day-Number Technique 155
3.35 Topocentric- Place Algorithm 160
3.4 DIFFERENTIAL ASTROMETRY 165
3.41 Virtual Place 165
3.42 Local Place 165
3.43 Astrometric Place 1 66
3.5 TRANSFORMATION TO FK5 SYSTEM AND EPOCH J2000.0 1 67
3.51 FK4 Zero-Point Correction in Right Ascension 167
3.52 The Correction to the FK4 Proper Motion System in Right Ascension 168
3.53 Elliptic Terms in Aberration 169
3.54 Precession 173
3.55 The Proper-Motion System 174
3.56 Equations for the Transformation of Catalogs from 81 950.0 to J2000.0 1 75
3.57 Transformation of Observational Catalogs 179
3.58 Numerical Examples 180
3.59 Matrix Method 180
3.6 REFERENCES 187
4 / TERRESTRIAL COORDINATES
AND THE ROTATION OF THE EARTH 1 99
4.1 INTRODUCTION 199
4.2 TERRESTRIAL COORDINATE SYSTEMS 200
4.21 The Figure of the Earth and the Reference Ellipsoid 200
4.22 Geocentric, Geodetic, and Astronomical Coordinates 202
4.23 Local Coordinate Systems 207
4.24 Geodetic Datums and Reference Systems 216
4.25 The Conventional Terrestrial Reference System 223
4.3 GRAVITY, THE TIDES, AND MOTIONS OF THE CRUST 224
4.31 Modeling the Earth's Gravity Field 224
4.32 A Representation of the Earth's Gravity Field 227
4.33 Solid Earth Tides 233
4.34 Ocean Tide Model 240
4.35 Site Displacement Due to Ocean and Atmospheric Loading 244
4.36 Plate Motions 249
4.37 Tidal Effects on UT1 250
CONTENTS
4.4 THE MONITORING OF THE ROTATION OF THE EARTH 251
4.41 Laser Ranging 254
4.42 Very Long Baseline Interferometry 255
4.43 Historical Methods 256
4.44 Alternative Techniques 258
4.45 International Services 262
4.5 DETERMINATION OF PAST VARIATIONS IN LENGTH OF DAY
AND THE POSITION OF THE POLE 265
4.51 Historical Variations in UT1 and Length of Day 265
4.52 Historical Variations in Polar Motion 269
4.6 REFERENCES 272
5 / ORBITAL EPHEMERIDES
OF THE SUN, MOON, AND PLANETS 279
5.1 FUNDAMENTAL EPHEMERIDES 279
5.11 Gravitational Model 280
5.12 The Ephemeris Reference Frame 280
5.13 The Astronomical Constants Used in the Ephemerides 280
5.2 COMPUTATION OF EPHEMERIDES 281
5.21 Mathematical Model 281
5.22 Numerical Integration 286
5.23 Orientation of Ephemerides 288
5.3 OBSERVATIONAL DATA FIT BY THE PLANETARY AND LUNAR
EPHEMERIDES 290
5.31 Optical Data 290
5.32 Radar-Ranging Data 294
5.33 Spacecraft Range Points 297
5.34 Lunar Laser Range Data 299
5.4 LEAST-SQUARES ADJUSTMENT OF THE EPHEMERIDES 300
5.41 The Observational Equations 301
5.42 The Solution Parameters 301
5.43 The Standard Deviations 303
5.5 NUMERICAL REPRESENTATION OF THE EPHEMERIDES 303
5.51 Chebyshev Polynomials 305
5.52 Chebyshev Coefficient Generation 305
5.53 Interpolation Error and Polynomial Degree 306
5.6 COMPUTATION OF OBSERVATIONAL EPHEMERIDES 307
5.61 Apparent Positions 308
5.62 Astro-metric Positions 308
5.63 Transit Ephemerides 308
5.7 ORBIT AND EPHEMERIDES OF OTHER BODIES
WITHIN THE SOLAR SYSTEM 31 0
5.71 Minor Planets and Comets 310
CONTENTS
Xi
5.8 KEPLERIAN ELEMENTS FOR THE POSITIONS OF THE MAJOR
PLANETS 315
5.9 BASIS FOR PRE-1984 EPHEMERIDES 317
5.91 Introduction of New Constants 317
5.10 REFERENCES 319
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 325
6.1 INTRODUCTION 325
6.11 Orbital Elements 325
6.1 2 Secular Perturbations of the Orbit 327
6.13 Perturbations due to Commensurabilities 330
6.14 Long-Period Perturbations by Other Satellites 332
6.15 Planetocentric Rectangular Coordinates 333
6.16 The Apparent Orbit 336
6.17 Calculating Tabulated Values 340
6.18 Notation 342
6.2 THE SATELLITES OF MARS 342
6.3 THE SATELLITES OF JUPITER 345
6.31 The Galilean Satellites 345
6.32 The Fifth Satellite, Amalthea 349
6.33 The Fourteenth Satellite, Thebe 351
6.34 The Sixth through Thirteenth Satellites 352
6.4 THE RINGS AND SATELLITES OF SATURN 354
6.41 The Rings of Saturn 354
6.42 The Satellites of Saturn 356
6.5 THE RINGS AND SATELLITES OF URANUS 368
6.6 THE SATELLITES OF NEPTUNE 373
6.61 Trilon 373
6.62 Nereid 375
6.7 THE SATELLITE OF PLUTO 377
6.8 REFERENCES 378
7 / PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS,
AND SATELLITES 383
7.1 INTRODUCTION 383
7.1 1 Rotational Elements and Cartographic Coordinates 383
7.12 Phases and Magnitudes 388
7.2 PHYSICAL EPHEMERIS OF THE SUN 397
7.3 PHYSICAL EPHEMERIS OF THE MOON 398
7.4 PHYSICAL EPHEMERIDES OF THE PLANETS 401
7.41 Mercury 401
7.42 Venus 402
xii
CONTENTS
7.43 Mars 403
7.44 Jupiter 403
7.45 Saturn 404
7.46 Uranus 405
7.47 Neptune 405
7.48 Pluto 406
7.5 PHYSICAL EPHEMERIDES OF THE SATELLITES 407
7.51 Satellites of Mars 407
7.52 Satellites of Jupiter 408
7.53 Satellites of Saturn 408
7.54 Satellites of Uranus 411
7.55 Satellites of Neptune 413
7.56 The Satellite of Pluto 414
7.6 PHYSICAL EPHEMERIDES OF THE ASTEROIDS 414
7.7 REFERENCES 417
8 / ECLIPSES OF THE SUN AND MOON 421
8.1 INTRODUCTION 421
8.11 Eclipse Data Available from the Nautical Almanac Office 422
8.12 Corrections to the Ephemerides 424
8.2 THE OCCURRENCE OF LUNAR AND SOLAR ECLIPSES 426
8.21 Overview 426
8.22 Geocentric Least Angular Separation 426
8.23 Occurrence of Lunar Eclipses 428
8.24 Occurrence of Solar Eclipses 431
8.3 SOLAR ECLIPSES 434
8.31 Fundamental Equations: Introduction 434
8.32 Besselian Elements 435
8.33 Coordinates of the Observer 441
8.34 Conditional and Variational Equations 446
8.35 Calculation of General Solar Eclipse Phenomena 450
8.36 Local Circumstances 461
8.4 LUNAR ECLIPSES 467
8.41 Introduction 467
8.42 Computations 467
8.5 TRANSITS 471
8.6 REFERENCES 472
9 / ASTRONOMICAL PHENOMENA 475
9.1 GENERAL ASPECTS OF THE NIGHT SKY 475
9.2 CONFIGURATIONS OF THE SUN, MOON, AND PLANETS 476
9.21 Interesting Phenomena of the Sun, Earth, and Moon 477
9.22 Geocentric Phenomena 478
9.23 Heliocentric Phenomena 481
CONTENTS
xiii
9.3 RISINGS, SETTINGS, AND TWILIGHT 482
9.31 Sunrise, Sunset, and Twilight 483
9.32 Moonrise and Moonset 485
9.33 Formulas Associated with Rising and Setting 486
9.34 Illumination 490
9.4 OCCULTATIONS 494
9.41 Occultations of Stars 494
9.42 Occultations of Planets 498
9.5 POLE-STAR TABLES 498
9.51 Derivation of the Pole Star Coefficients 501
9.6 REFERENCES 502
1 0 / STARS AND STELLAR SYSTEMS 505
1 0.1 SOURCES OF DATA ON STARS AND STELLAR SYSTEMS 505
10.11 Compiled Catalogs of Stellar Positions and Motions 506
10.12 Standard Reference Catalogs 507
10.13 Observational Positional Catalogs 508
10.14 Other Catalogs and Lists 509
10.15 Data Centers and Their Facilities 509
10.2 STELLAR DATA IN THE ASTRONOMICAL ALMANAC 509
10.21 Bright Stars 512
10.22 Variable Stars 513
10.23 Double and Multiple Stars 515
10.24 Photometric Standards 516
10.25 Radial-Velocity Standards 518
10.26 Spectral Classification 519
10.27 Pulsars 521
10.3 CLUSTERS AND GALAXIES
10.31 Open Cluster Data 522
10.32 Globular Star Cluster Data
10.33 Bright Galaxies 528
10.34 Quasi-Stellar Objects 530
10.4 SOURCES CATEGORIZED BY WAVELENGTH REGION 531
10.41 Radio-Source Positional Calibrators 531
10.42 Radio-Flux Calibrators 532
10.43 X-Ray Sources 533
10.5 REFERENCES 534
1 1 / COMPUTATIONAL TECHNIQUES 541
11.1 INTRODUCTION TO COMPUTING TECHNIQUES 541
11.2 INTERPOLATION AND SUBTABULATION 546
1 1 .21 Introduction and Notation 546
11.22 Interpolation Formulas 547
1 1 .23 Inverse Interpolation 548
11.24 Polynomial Representations 548
522
526
xiv
CONTENTS
11.3 PLANE AND SPHERICAL TRIGONOMETRY 549
11.4 MATRIX AND VECTOR TECHNIQUES 552
1 1 .41 Rotation of Axes Using Matrices 552
11.42 Spherical Coordinates Using Vectors 553
1 1 .43 Specific Coordinate Transformations 555
11.5 NUMERICAL CALCULUS 560
11.51 Numerical Differentiation 560
1 1 .52 Numerical Integration 562
11.6 STATISTICS 566
1 1 .61 The Accumulation of Error 566
1 1 .62 The Method of Least Squares 568
11.7 REFERENCES 574
12 /CALENDARS 575
12.1 INTRODUCTION 575
12.11 Astronomical Bases of Calendars 576
12.12 Nonastronomical Bases of Calendars: the Week 577
12.13 Calendar Reform and Accuracy 578
12.14 Historical Eras and Chronology 579
12.2 THE GREGORIAN CALENDAR 580
12.21 Rules for Civil Use 580
12.22 Ecclesiastical Rules 581
1 2.23 History of the Gregorian Calendar 583
12.3 THE HEBREW CALENDAR 584
12.31 Rules 584
12.32 History of the Hebrew Calendar 588
12.4 THE ISLAMIC CALENDAR 589
12.41 Rules 589
12.42 History of the Islamic Calendar 591
12.5 THE INDIAN CALENDAR 591
12.51 Rules for Civil Use 591
12.52 Principles of the Religious Calendar 592
12.53 History of the Indian Calendar 594
12.6 THE CHINESE CALENDAR 594
12.61 Rules 595
12.62 History of the Chinese Calendar 599
12.7 JULIAN DAY NUMBERS AND JULIAN DATE 600
12.8 THE JULIAN CALENDAR 600
12.81 Rules 601
12.82 History of the Julian Calendar 601
12.9 CALENDAR CONVERSION ALGORITHMS 603
12.91 Converting Day of the Week 603
12.92 Converting between Gregorian Calendar Date and Julian Day Number 604
CONTENTS
XV
12.93 Converting Between Islamic Tabular Calendar Date
and Julian Day Number 604
12.94 Converting between Indian Civil Calendar and Julian Day Number 605
12.95 Converting between Julian Calendar Date and Julian Day Number 606
12.10 REFERENCES 606
13 / HISTORICAL INFORMATION 609
13.1 HISTORY OF THE ALMANACS 609
13.11 The American Ephemeris 609
13.12 The American Ephemeris and Nautical Almanac 613
13.13 The Cooperative British and American Almanacs (1960 to Present) 614
13.2 HISTORY OF INTERNATIONAL COOPERATION 616
13.21 Cooperation Prior to the IAU 616
13.22 International Astronomical Union 617
13.23 Other International Organizations 621
13.3 HISTORICAL LIST OF AUTHORITIES 621
13.31 Introduction 621
1 3.32 The Nautical Almanac (1 767-1 900) 622
13.33 The American Ephemeris (1855-1900) 631
13.34 The Nautical Almanac, and The American Ephemeris (1901-1 983) 639
1 3.35 System of Constants (1 968-1 983) 656
13.4 REFERENCES 657
1 4 / RELATED PUBLICATIONS 667
14.1 CURRENT PUBLICATIONS 667
14.1 1 Joint Publications of the Royal Greenwich Observatory and the United States
Naval Observatory 667
14.12 Other Publications of the United States Naval Observatory 668
14.13 Other Publications of the Royai Greenwich Observatory 669
14.14 Publ ications of Other Countries 669
14.2 ASTRONOMICAL PAPERS PREPARED FOR THE USE OF THE AMERICAN
EPHEMERIS AND NAUTICAL ALMANAC 670
14.3 UNITED STATES NAVAL OBSERVATORY CIRCULARS 675
14.4 PUBLICATIONS OF THE UNITED STATES NAVAL OBSERVATORY,
SECOND SERIES 677
14.5 SELECTED NAO TECHNICAL NOTES OF THE ROYAL GREENWICH
OBSERVATORY 684
14.6 LISTS OF APPENDICES AND SUPPLEMENTS 685
14.61 The British Nautical Almanac 685
14.62 The American Ephemeris 688
14.63 Joint Supplements and Appendices 691
xvi
CONTENTS
15/ REFERENCE DATA 693
15.1 Fundamental Constants (1986 Recommended Values) 693
15.2 IAU (1976) System of Astronomical Constants 696
1 5.3 Time and Standard Epochs 698
15.4- Sun, Earth, and Moon 700
15.5 Geodetic Reference Systems 702
15.6 Planets: Mean Elements 704
15.7 Planets: Rotational Data 705
15.8 Planets: Physical and Photometric Data 706
15.9 Satellites: Orbital Data 708
15.10 Satellites: Physical and Photometric Data 710
15.11 Planetary Rings 712
15.12 Constellation Names and Abbreviations 713
15.13 Mathematical Constants 714
15.14 Energy Conversion Factors 715
15.15 Units of Length, Speed, and Mass 716
15.16 Greek Alphabet 717
15.17 International System of Units (SI) 718
15.1 REFERENCES 719
GLOSSARY 721
Index 741
LIST OF FIGURES
1 .23 1 . 1 Calculation of sidereal time 5
1.233.1 Variation in the equation of time through the year 6
1.31.1 Representation of the vector r in rectangular coordinates 10
1.321.1 Equatori al and ecliptic reference planes 1 2
1,333.1 Geocentric and geodetic coordinates 15
1 .337. 1 Relation between geographic latitude and the altitude of the celestial pole 1 8
1 .36 1 . 1 Parallax of an object 22
1 .4 1 1 . 1 Geometric properties of conic sections 25
1 .4 1 2. 1 Angular orbital elements 27
2.25.1 Meridian relations and time 55
2.27.1 World map of time zones 57
3.21.1 The ecliptic and equator at epoch and date 100
3.21.2 The precession angles Q,, za- and 6A 102
3.222.1 The mean and true equators of date 115
3.251.1 Light-time aberration 128
3.252.1 Stellar aberration 128
3.26.1 Gravitational light deflection 136
3.26.2 Light from the planets and stars deflected by the sun 1 38
4.21. 1 The Earth's surface and reference surfaces 201
4.22. 1 Geocentric and geodetic latitude 202
4.22.2 Astronomic latitude and longitude 203
4.22.3 Deflection of the vertical on a unit sphere 204
4.22.4 Geocentric cartesian coordinates 205
4.231.1 Altitude and azimuth 208
4.233. 1 UTM and UPS grid zone designations 21 1
4.242,1 Major geodetic datum locations 219
4.5 1 . 1 Length of day from 1 656 to 1 988 267
4.5 1 .2 At from A.D. 700 to 1 600 268
4.5 1.3 At from 700 B.C. to A.D. 200 268
4.5 1 .4 log At from 700 B.C. to a.d. 1 980 269
4.52. 1 Polar motion, 1 980 September 28 to 1 990 July 27 270
4.52.2 X and Y components of polar motion from the ILS, 1 899.9 to 1979.0 27 1
4.52.3 X and K components of polar motion, 1846.0 to 1891.5 272
5.212.1
5.321.1
The coordinate system, in which figure-induced accelerations are calculated 283
Diagram of geometry for planetary radar ranging 295
xvii
xvjjj LIST OF FIGURES
5.714. 1 The relationship between the B 1950.0 and J2000.0 reference frames and the orbital
plane 314
6.11.1 The orbital elements used to describe the orbital plane relative to a reference plane 326
6.1 1.2 An equivalent form of the orbital elements. The end-on view of a great circle on the
celestial sphere is represented as a straight line. 326
6.12.1 The Laplacian plane 328
6.15.1 The satellite orbit referred to an intermediate plane 334
6.15.2 Coordinates of the satellite relative to the planet 335
6.16.1 Planetocentric celestial sphere 337
6. 16.2 Spherical triangle used to compute U, B, and P 337
6. 1 6.3 Spherical triangle used to compute a and p — P 337
6. 16.4 Apparent orbit of a satellite as projected on the geocentric sphere 339
6.2. 1 Reference system for Phobos and Deimos 344
6.32. 1 Reference system for Amalthea and Thebe 350
6.421.1 Reference plane for Mimas, Enceladus, Tethys, and Dione 357
6.422.1 Reference system for Rhea, Titan, Hyperion, Iapetus, and Phoebe 361
6.422.2 Angles needed to evaluate solar perturbations on Titan 362
6.5.1 Reference system for the Uranian satellites 372
6.61.1 Reference system for Triton 374
6.62.1 Reference system for Nereid 376
7.11.1 The position of the north pole and prime meridian of a planet in Earth equatorial
coordinates 384
7. 1 1.2 Planetocentric and planetographic coordinate systems 386
7.12.1 The basic vectors and the angle of illumination 389
7.12.2 The geometric appearance of the apparent polar radius, b' 391
7. 1 2.3 The disk of a planet as seen by an observer on the Earth 392
7.12.4 Planetocentric unit vectors for points of interest on the disk of the planet, and the angles
between the Earth's equator of J2000.0 and the planet's equator of date 394
8.22.1 Geometric construct for determining whether eclipse will occur 426
8.231 . 1 Geometric parameters of the Earth's penumbral shadow 428
8.231.2 Geometric parameters of the Earth's umbral shadow 429
8.2321.1 Limiting conditions for lunar eclipses 430
8.242. 1 Geometric parameters for a partial solar eclipse, when Moon is externally tangent 432
8.2421 .1 Geometric parameters for a partial solar eclipse when Moon is internally tangent 434
8.32 1 . 1 Transformation of geocentric equatorial coordinates to the fundamental plane 435
8.323. 1 Components of shadow cones in the Besselian fundamental reference system 438
8.323.2 Relationships among angular semidiameter, apparent semidiameter, and parallax 439
8.323.3 Vertex angle of the penumbra 440
8.323.4 Vertex angle of the umbra 440
8.331.1 Angular quantities in the geocentric equatorial plane 442
8.352.1 Relationship between rectangular and polar reference systems 451
8.3556.1 Definition of auxiliary quantity ip 458
8.3623. 1 Magnitude is the fraction of the linear diameter of the Sun covered by the Moon 464
8.3623.2 Obscuration is the fraction of the area of the solar disk obscured by the Moon 465
9.311.1
The analemmic curve 485
LIST OF FIGURES
9.331.1 The horizon at rising or setting 488
9.34.1 Ground illumination from various sources 491
9.41.1 Fundamental plane showing path of occultation 497
9.5.1 PZS triangle for Polaris 499
9.51.1 Polaris north polar distance 2096-2 1 05 502
1 1 . 1 . 1 Spherical triangle 542
11.3.1 Arc of circle 551
11.42.1 Triangle on unit sphere 555
LIST OF TABLES
1 .342. 1 Variations in the Earth's Rate of Rotation 19
2,33.1 Techniques for Measuring Earth's Rotation 62
2,45. 1 Error Budgets for Common View GPS Time Transfer 68
2.58. 1 Time and Frequency Steps by WWV and MC (USNO) 86
3.211.1 Accumulated Precession Angles 104
3.214.1 Precession Angles 1984 January ld0h 107
3.222. 1 Nutation in Longitude and Obliquity Referred to the Mean Ecliptic of Date 1 1 2
3.222.2 Fundamental Arguments 114
3.224.1 Corrections to IAU 1980 Nutation Series 1 16
3.224.2 Planetary Terms in Nutation, Combined Direct and Indirect Effects 1 1 8
3.253.1 Coefficients (nma I c) for the Major Planets 132
3.26.1 Apparent Deflection Angles 138
3.344.1 Second-Order Terms 159
3.344.2 Errors Due to Neglecting Second-Order Terms 1 60
3.58. 1 Selected Star Positions on FK4 and FK5 Systems 1 8 1
4.242.1 Earth Ellipsoids 220
4.242.2 Geodetic Datums 221
4.32.1 GEM-T1 Normalized Coefficients (x 106) 228
4.331.1 Step 2 Solid Tide Corrections When k2 = 0. 30 in Step 1 (Using a Cutoff Amplitude of
9x I0~n for Am6ksHs) 235
4.34.1 Ocean Tide Coefficients from the Schwiderski Model 242
4.351.1 Displacement Due to Ocean Loading (cm in amplitude and degree in phase) 245
4.36.1 Cartesian Rotation Vector for Each Plate Using Kinematic Plate Model AMO-2 (No Net
Rotation) 250
4.37. 1 Zonal Tide Terms with Periods Up to 35 Days 252
4.37.2 Zonal Tide Terms with Periods Greater than 35 Days 253
5.214.1 Lunar Libration Angles and Rates 285
5.311.1 Transit Circle Observations from the U.S. Naval Observatory that have been used in the
JPL Ephemerides, DEI 18 29 1
5.322. 1 Radar-Ranging Observations Used Directly in DEI 18 297
5.332.1 Mariner 9 Orbiter Normal Range Points to Mars 298
5.41.1 The Observational Data Used for the Adjustment of DE118/LE62 301
5.42.1 Values and Formal Standard Deviations of the Constants Used in DEI 18/ LEI 18
(DE200/LE200) 302
5.42.2 The Initial Conditions of the Ephemerides at JED 2440400.5 in AU and AU/day 304
5.53. 1 Granule Length and Polynomial Degree for the 1 1 Ephemeris Bodies 307
xxi
xxii
5.8.1
5.8.2
6.13.1
7.41.1
7.42.1
7.43.1
7.44.1
7.45.1
7.46.1
7.47.1
7.48.1
7.51.1
7.52.1
7.52.2
7.53.1
7.53.2
7.54.1
7.55.1
7.56.1
8.113.1
8.422.1
9.211.1
9.22.1
9.22.2
9.221.1
9.222.1
9.222.2
9.23.1
9.34.1
1 1 I.I
11.1.2
I 1.1.3
11.21.1
11.3.1
11.51.1
1 1.61.1
12.21.1
12.31.1
12.31.2
12.31.3
12.311.1
LIST OF TABLES
Classical Keplerian elements at the epoch J2000 (JED 2451545.0) 316
Approximate maximum errors of the Keplerian formulas over the interval
1800-2050 316
The Principal Commensurabilities Among Satellites 332
Physical Ephemeris Parameters for Mercury 402
Physical Ephemeris Parameters for Venus 402
Physical Ephemeris Parameters for Mars 403
Physical Ephemeris Parameters for Jupiter 404
Physical Ephemeris Parameters for Saturn 404
Physical Ephemeris Parameters for Uranus 405
Physical Ephemeris Parameters for Neptune 406
Physical Ephemeris Parameters for Pluto 406
Rotation Parameters for Mars' Satellites 407
Rotation Parameters for Jupiter's Satellites 409
Standard Cartographic Longitudes for Jupiter's Satellites 409
Rotation Parameters for Saturn's Satellites 410
Standard Cartographic Longitudes for Saturn's Satellites 41 1
Rotation Parameters for Uranus' Satellites 412
Rotation Parameters for Neptune's Satellites 413
Rotation Parameters for Pluto's Satellite 414
U.S.N.O. Solar Eclipse Circulars 424
Sequences and Conditions for Contact Times 470
Time of Commencement of the Seasons 477
Geocentric Phenomena for which /(r) = 0 479
Geocentric Phenomena for which f'(t) = 0 479
Visibility Criteria for Geocentric Phenomena 480
Synodic Periods of the Planets and First Asteroids 480
Long-Period Cycles of Mercury. Venus, and Mars 48 1
Heliocentric Phenomena 482
Coefficients for Calculating Ground Illumination 492
Precision of Angle and Number of Decimals for Trigonometric Functions 544
The Method of Inverse Use 545
Range of Precision of the Inverse Determination of an Angle 545
Differences in Tabular Arguments 547
Formulas for Plane and Spherical Triangles 550
Derivatives to an Order of 10 561
Accumulation of Error in Arithmetical Operations 567
Months of the Gregorian Calendar 58 1
Classification of Years in the Hebrew Calendar 584
Months of the Hebrew Calendar 585
Terminology of the Hebrew Calendar 585
Lunation Constants for Determining Tishri 1 586
LIST OF TABLES XXIH
12.41.1 Months of Tabular Islamic Calendar 590
12.51.1 Months of the Indian Civil Calendar 592
12.52.1 Solar Months of the Indian Religious Calendar 593
12.61.1 Chinese Sexagenary Cycle of Days and Years 595
12.61.2 Chinese Solar Terms 597
12.82.1 Roman Dating in the Julian Calendar 602
1 3,348. 1 Adopted Values for Coefficients of Nutation 654
15. 1 Fundamental Constants ( 1 986 Recommended Values) 693
15.2 IAU (1976) System of Astronomical Constants 696
15.3 Time and Standard Epochs 698
15.4 Sun, Earth, and Moon 700
15.5 Geodetic Reference Systems 702
15.6 Planets: Mean Elements 704
15.7 Planets: Rotational Data 705
15.8 Planets: Physical and Photometric Data 706
15.9 Satellites: Orbital Data 708
15.10 Satellites: Physical and Photometric Data 710
15.11 Planetary Rings 712
15.12 Constellation Names and Abbreviations 713
15.13 Mathematical Constants 714
15.14 Energy Conversion Factors 715
15.15 Units of Length, Speed, and Mass 716
15.16 Greek Alphabet 717
15.17 International System of Units (SI) 718
Foreword
The Explanatory Supplement to the Astronomical Ephemeris and the American
Ephemeris and Nautical Almanac was first published in 1961 "to provide the user
of these publications with fuller explanations of these publications themselves,"
This supplement was reprinted with amendments in 1972, 1974, and 1977. It was
allowed to go out of print because the International Astronomical Union decided to
introduce new astronomical constants, a new standard epoch and equinox, new time
arguments, a new astronomical reference frame, and new fundamental ephemerides,
all of which required major revisions to the supplement. In addition, The Astro-
nomical Ephemeris and The American Ephemeris and Nautical Almanac series were
continued from 1981 with a new title, The Astronomical Almanac, which contains a
revised content and arrangement and is printed only in the United States. The work
of computation, preparation, proofreading, and production of reproducible material
is still shared between the United Kingdom and the United States of America.
Many changes in the arrangement of the almanac were introduced in the edition
for 1981, and major changes in the basis of the ephemerides were made in the edition
for 1984. The changes in 1981 included: the replacement of the hourly ephemeris
of the Moon by a tabulation of daily polynomial coefficients; the introduction of
a new system of rotational elements for the planets; the extension of the scope of
the data on satellites; the inclusion of orbital elements and other data for minor
planets of general interest; the extension of the list of bright stars; the inclusion of
new lists of data for other types of stars and nonstellar objects; new explanatory
material; and a glossary of terms. The changes in 1984 included: the replacement
of the classical theories by the Einsteinian theories of special and general relativ-
ity; the replacement of ephemeris time by dynamical timescales; the adoption of
new fundamental heliocentric ephemerides based on a numerical integration of the
motion of all the planets and of the Moon; and the use of the IAU (1976) system
XXV
xxv| FOREWORD
of astronomical constants, the standard equinox of J2000.0, and the FK5 celestial
reference system. An account of these changes was given in the 39-page Supplement
to the Astronomical Almanac for 1984, which was bound with the Almanac, and is
also given here in great detail in Chapter 13.
Most of the text in this supplement has been written for readers who are familiar
with the principal concepts of spherical and dynamical astronomy but who require
detailed information about the data published in The Astronomical Almanac and
about how to use the data for particular purposes. Similarly, the reference data
given in this supplement are presented in forms that are appropriate to users who
understand the significance of the quantities whose values are given. To a large
extent, the chapters are independent of each other, but an introductory overview
has been given in Chapter 1 and a glossary of terms has been given at the end of
the volume. References to textbooks and other sources of background information
are given at the end of each chapter.
Preliminary proposals for the new edition of the Explanatory Supplement were
drawn up in 1979, and more detailed outlines of the Supplement were prepared in
1986 by the staffs of Her Majesty's Nautical Almanac Office and the U.S. Nautical
Almanac Office. By 1988, it was evident that Her Majesty's Nautical Almanac
Office would not be able to participate as originally planned, and the U.S. Nautical
Almanac Office took over the entire project.
The supplement is organized by chapters and sections such that it can be up-
dated in the future. It is planned that future reprints will incorporate developments
and improvements. We hope that this new publication will prove to be even more
useful than its predecessor.
Preface
The primary purpose of this revised Explanatory Supplement is to provide users of
The Astronomical Almanac with more complete explanations of the significance,
sources, methods of computation, and use of the data given in the almanac than
can be included annually in the almanac itself. The secondary purpose is to provide
complementary information that doesn't change annually, such as conceptual ex-
planations, lists of constants and other data, bibliographic references, and historical
information relating to the almanac. It is hoped that the Explanatory Supplement
will be a useful reference book for a wide range of users in the fields of astronomy,
geodesy, navigation, surveying, and space sciences, and also teachers, historians,
and people interested in the field of astronomy.
Many users of the almanac are not the professional astronomers for whom it
is primarily designed, and so this supplement contains some explanatory material
at an elementary level; it is not, however, intended for use as a basic textbook
on spherical and dynamical astronomy. In some respects it does supplement such
textbooks since it is concerned with new concepts or new techniques.
This supplement differs in many respects from its predecessor, the Explanatory
Supplement to The Astronomical Ephemeris and The American Ephemeris and
Nautical Almanac. Vector and matrix notations have been introduced and more
diagrams have been provided. Simple conversion tables and tables of quantities
that can be calculated directly from simple formulas have been omitted. Detailed
step-by-step examples have been omitted, and approximation methods have not
been given. Most of the text is new but historical material has been carried over for
the convenience of those who do not have ready access to the previous supplement.
This supplement has been prepared by the Nautical Almanac Office of the
United States Naval Observatory. Material has been contributed by scientists from
the Nautical Almanac Office of the Royal Greenwich Observatory, Jet Propulsion
xxvii
xxviii
PREFACE
Laboratory, Bureau des Longitudes, Time Service and Astrometry Departments
of the U.S. Naval Observatory, and other scientists. The authors of each chapter
have been indicated, but other individuals may have been involved in contributing,
improving, and checking the material. The valuable assistance that has been given
in many ways by other astronomers and scientists is gratefully acknowledged.
Suggestions for improvement of this supplement, and of The Astronomical Al-
manac itself would be welcomed. They should be sent to the Director, Orbital
Mechanics Department, U.S. Naval Observatory, Washington, DC 20392.
James B. Hagan
Captain, U.S. Navy
Superintendent, U.S. Naval Observatory
EXPLANATORY SUPPLEMENT
TO THE ASTRONOMICAL ALMANAC
CHAPTER
Introduction to Positional Astronomy
by P.K. Seidelmann and G.A. Wilkins
1.1 INTRODUCTION
1.11 Purpose
The Astronomical Almanac gives data on the positions and, where appropriate,
orientations of the Sun, Moon, planets, satellites, and stars as they may be seen
from the surface of the Earth during the course of a year. A proper appreciation
of the significance of these data requires a basic understanding of the concepts of
spherical astronomy, which explain how the varying directions of celestial objects
may be represented by positions on the surface of the celestial sphere. In addition, an
appreciation of why these celestial objects appear to move in the ways predicted in
The Astronomical Almanac requires an understanding of the concepts of dynamical
astronomy, which provides a mathematical explanation of the objects' motions in
space under the influence of their mutual gravitational attractions. Spherical and
dynamical astronomy together form what is referred to here as positional astronomy.
This text has been written for readers familiar with the principal concepts of
spherical and dynamical astronomy who require detailed information about the
computation and use of the data published in The Astronomical Almanac. The
primary purpose of this introductory chapter is to introduce the concepts, termi-
nology, and notation that are used throughout The Astronomical Almanac and this
supplement; rigorous definitions, formulas, and further explanatory information are
given in the later chapters of this supplement. The glossary gives concise definitions
of words particular to spherical and dynamical astronomy. The reference data are
presented in forms that are appropriate to users who understand the significance
of the quantities whose values are given.
1
2
1 .2 TIMESCALES AND CALENDARS
EXPLANATORY SUPPLEMENT
1.21 Atomic Timescales
1.211 International Atomic Time For scientific, practical, and legal purposes the
standard unit for the measurement of intervals of time is the SI second, which is
defined by the adoption of a fixed value for the frequency of a particular transition of
cesium atoms. Time can be measured in this unit by the use of time standards based
on processes of physics. Cesium frequency standards, hydrogen masers, ion storage
devices, and other such devices are able to count seconds and subdivide them very
precisely. Thus, such a device can provide a timescale whose accuracy is dependent
on the precision of the measurement and the stability of the device. Such a timescale
provides a measure of time for identifying the instants at which events occur; the
interval of time between two events can be calculated as the differences between
the times of the events. The results of the intercomparison of about 200 frequency
standards located around the world are combined to form a standard timescale
that can be used for identifying uniquely the instants of time at which events
occur on the Earth. This standard timescale is known as International Atomic
Time (TAI). It is the basis for all timescales in general use. It is distributed by
many different means, including radio time signals; navigation systems, such as the
Global Positioning System, LORAN-C, and OMEGA; communication satellites;
and precise time standards.
1.212 Relativistic Effects In high- precision timekeeping, and for some purposes
in solar-system dynamics and astrometry, it is necessary to take into account the
effects of special and general relativity and to recognize, for example, that the rate of
an atomic clock depends on the gravitational potential in which it is placed and that
the rate will appear to depend on its motion relative to another clock with which
it is compared. In particular, one should recognize that the independent variable,
or timescale, of the equations of motion of the bodies of the solar system (or of a
subset of them) depends upon the coordinate system to which the equations refer.
The relationship between any such timescale and TAI, which is appropriate for use
at sea- level on the surface of the Earth, may be specified by an appropriate formula
containing periodic terms and an arbitrary linear term. Two such timescales have
been given special names, and these have been used in The Astronomical Almanac
since 1984. Terrestrial Dynamical Time (TDT) is used as the timescale for the
geocentric ephemerides (giving, for example, apparent positions with respect to
the center of the Earth), whereas Barycentric Dynamical Time (TDB) is used as
the timescale for the ephemerides that refer to the center of the solar system.
TDT differs from TAI by a constant offset, which was chosen arbitrarily to give
continuity with ephemeris time (see Section 1.22), whereas TDB and TDT differ
1 / INTRODUCTION TO POSITIONAL ASTRONOMY 3
by small periodic terms that depend on the form of the relativistic theory being
used.
In 1991 the International Astronomical Union (IAU) adopted resolutions in-
troducing new timescales which all have units of measurement consistent with the
proper unit of time, the SI second. Terrestrial Time (TT) is the time reference
for apparent geocentric ephemerides, has a unit of measurement of the SI second,
and can be considered equivalent to TDT. Geocentric Coordinate Time (TCG) and
Barycentric Coordinate Time (TCB) are coordinate times for coordinate systems
that have their origins at the center of mass of the Earth and of the solar system,
respectively. Their units of measurement are chosen to be consistent with the proper
unit of time, the SI second. Because of relativistic transformations, these timescales
will exhibit secular variations with respect to Terrestrial Time. For example, TDB
and TCB will differ in rate by approximately 49 seconds per century.
1.22 Dynamical Time
The equations of motion of the bodies of the solar system involve time as the inde-
pendent variable. Until 1960, mean solar time was used, but when it was recognized
that the rotation of the Earth was irregular, a new timescale was introduced that
corresponded to the independent variable. This was called ephemeris time (ET) and
was based on the motion of the Sun. Later, when it was necessary to distinguish be-
tween relativistic effects — which cause differences between timescales for the center
of the Earth and the center of the solar system — the dynamical time arguments,
Terrestrial Dynamical Time, and Barycentric Dynamical Time, were introduced.
TDT was defined in a way that maintains continuity with ET. Since ET was not
specified as either TDT or TDB, either can be considered to be the extension of ET.
Since TDT is defined in terms of TAI, which can be determined only back to 1956,
the determination of dynamical time prior to 1956 must be based on comparison
of observations and theories of the motions of the Sun, Moon, and planets.
1 .23 Rotational Timescales
As the Earth rotates about its axis, it also moves in its orbit around the Sun.
Thus, while the Earth rotates once with respect to a fixed star, the Earth moves
in its orbit so that additional rotation is necessary with respect to the Sun. The
rotation of the Earth with respect to the equinox is called sidereal time (ST). The
Earth's rotation with respect to the Sun is the basis of Universal Time, also called
solar time. Since the rotation of the Earth is subject to irregular forces, sidereal
time and Universal Time are irregular with respect to atomic time. Sidereal time is
the hour angle of the catalog equinox and is subject to the motion of the equinox
itself due to precession and nutation. Ot herwise, it is a direct measure of the diurnal
rotation of the Earth. Sidereal time reflects the actual rotation of the Earth and can
4 EXPLANATORY SUPPLEMENT
be determined by observations of stars, artificial satellites, and extragalactic radio
sources. On the other hand, the apparent diurnal motion of the Sun involves both
the nonuniform diurnal rotation of the Earth and the motion of the Earth in its
orbit around the Sun. In practice, Universal Time is directly related to sidereal time
by means of a numerical formula. For each local meridian there is a corresponding
local sidereal time. The measure of the rotation of the Earth with respect to the
true equinox is called apparent sidereal time. The measure with respect to the mean
equinox of date is referred to as mean sidereal time. Apparent sidereal time minus
mean sidereal time is called the equation of the equinoxes.
1.231 Sidereal Time The local hour angle of the equinox has a special significance
since it serves to specify the orientation of the celestial (equatorial) coordinate
system with respect to the local terrestrial coordinate system in which hour angle
is measured. The local hour angle of the equinox is known as the local sidereal time
(LST) and it increases by 24" in a sidereal day. The fundamental relation for some
celestial object X is
local hour angle of X = local sidereal time — right ascension of X. (1.231-1)
This may be written in the abbreviated form
LHA X = LST - RA X. (1.231-2)
The value of the local sidereal time is equal to the right ascension of the local
meridian, and so may be determined by observing the meridian transits of stars
of known right ascensions, the positions of radio sources using Very Long Baseline
Interferometry, or the distance from a location on Earth to a retrorefiector on the
Moon by Lunar Laser-Ranging. Local sidereal time can be calculated from the
sidereal time on the prime meridian (Greenwich) when the difference in geographic
longitude is known (see Figure 1.231.1); thus
local sidereal time = Greenwich sidereal time + east longitude. (1.231-3)
1.232 Solar and Universal Time The general form of the relationship between
sidereal time and Universal Time may be derived by substituting the Sun for X in
the above equation. The local hour angle of the Sun is, by definition, 12 hours less
than local apparent solar time (LAT) and so
LAT = LST — RA Sun - 12? (1.232-1)
The right ascension of the Sun does not vary uniformly with time nor does the
Sun move on the equator, but it is possible to introduce the concept of a point U
1 / INTRODUCTION TO POSITIONAL ASTRONOMY
Local meridian
Computed Mean Sun (U)
Equinox
1.002738 AT
GHA X
Equation of time
DAT]
KA U
1
UT.
GHA Sun
LMT -
RA Sun
LA Sun
>
s
GST
Longitude
LST
LHA X
RA X
1 r
Greenwich meridian X
True Sun
Figure 1.231.1
Calculation of sidereal time
that moves around the celestial equator at a uniform rate. Hence, Universal Time
is defined in terms of Greenwich sidereal time (GST) by an expression of the form
UT = GST - RA U- 12\
(1.232-2)
where the coefficients in the expression for RA U are chosen so that UT may for
most purposes be regarded as mean solar time on the Greenwich meridian.
1.233 Equation of Time The difference between local mean time (LMT) and local
apparent solar time is known as the equation of time, and the relationship is now
expressed in the form
LAT = LHA Sun + 12h = LMT + equation of time,
(1.233-1)
although the equation of time used to be regarded as the correction to be applied to
apparent time to obtain mean time. The principal contributions to the equation of
time arise from the eccentricity of the Earth's orbit around the Sun (which causes
a nonuniformity in the apparent motion of the Sun around the ecliptic) and the
inclination of the plane of the ecliptic to the plane of the equator. The equation of
time varies through the year in a smoothly periodic manner by up to 16 minutes,
as shown in Figure 1.233.1
6
EXPLANATORY SUPPLEMENT
+3 m 45 s
+2 m 43 s
OmOs
— 3 rn 16s
-4 m 27 s
-9 m 12 s
-12m45s
-13 m 24 s
-14m 14s
+16m 19s
+15 m 30s
+ 14m Os
+1 1 m 36 s
+9 m 43 s
+5m 13 s
+4m 29 s
+2 m 33 s
-0 m 35 s
-2m 39s
-3 m 23 s
-4 m 36 s
-5 m 47 s
— 6 m 21 s
Figure 1.233.1
Variation in the equation of time through the year
1 .24 Coordinated Universal Time (UTC)
Although TAI provides a continuous, uniform, and precise timescale for scientific
reference purposes, it is not convenient for general use. In everyday life it is more
convenient to use a system of timescales that correspond to the alternation of day
and night, apply over fairly wide areas, and can be easily related to each other and
to TAI. In these timescales, the numerical expression, or measure, of the time of
an event is given in the conventional form of years, months, days, hours, minutes,
seconds, and decimals of seconds — i.e., as a calendar date and time of day. The
standard time on the prime meridian is known as Coordinated Universal Time
(UTC). UTC is an atomic timescale that is kept in close agreement with Universal
Time (UT), which is a measure of the rotation of the Earth on its axis. The rate
of rotation of the Earth is not uniform (with respect to atomic time), and the
difference between TAI and UT is increasing irregularly by about 1 second every
1 / INTRODUCTION TO POSITIONAL ASTRONOMY 7
18 months. The difference between UTC and TAI is always an integral number of
seconds. UTC is maintained in close agreement, to better than one second, with UT
by introducing extra seconds, known as leap seconds, to UTC, usually at the end of
the last day of June or December. The Earth is divided into standard-time zones
in which the time kept is that of a standard meridian (multiples of 15° longitude).
Thus, the local noon at any place in the zone is near twelve noon of the standard
time. These standard times usually differ from UTC by integral numbers of hours.
In summer, the time may be advanced to increase the hours of daylight in the
evening.
1.241 Greenwich Mean Time (GMT) In the past, the term Greenwich Mean Time
(GMT) has been used. It is the basis of the civil time for the United Kingdom and,
as such, is related to UTC. However, in navigation terminology, GMT has been used
as Universal Time. For precise purposes it is recommended that the term GMT not
be used, since it is ambiguous.
Prior to 1925, GMT was measured for astronomical purposes from noon to
noon, so that the date would not change in the middle of a night for an observer in
Europe. In 1925 that practice was discontinued, and GMT was then measured from
midnight to midnight. Thus, care must be taken in using time references before
1925.
1.25 The Enumeration of Dates
1.251 Civil Calendars The alternation of day and night is a clear physical phe-
nomenon that is repetitive and countable; so the solar day is the basic unit of all
calendars. Some calendars use the lunar month and/or solar year for longer units,
but these periods are neither fixed nor made up of integral numbers of days. More-
over, the length of the lunar month is not a simple rational fraction of the solar
year. However, the cycle of weeks, each of seven named days, is very widely used,
and it continues independently of the enumeration of days in the calendar.
The Gregorian calendar is now commonly used throughout most countries of the
world for the identification of solar days. It was derived from the Julian calendar,
so named because its system of months was introduced by Julius Caesar. The
present system for the enumeration of years in the Julian calendar was introduced
only gradually after A.D. 500; the extension of the present system to earlier years
is known as the Julian proleptic calendar. The change from the Julian calendar
to the Gregorian calendar took place in most countries in western Europe (and
in their colonies) between 1583 and 1753. However, many other countries did not
change systems until much later, and at times both systems were used concurrently.
One must be very careful when using astronomical data published at times and in
places where the calendrical system in use may be in doubt. Many other calendars
have been used, and may still be used for some purposes, in other countries. Such
g EXPLANATORY SUPPLEMENT
calendars often depend on various astronomical phenomena and cycles, but are now
primarily of religious significance.
A particular instant of time may be identified precisely by giving the calendar
date (assumed to be Gregorian unless otherwise specified) and the time of day in a
specified time system.
1.252 Julian Date For many astronomical purposes it is more convenient to use a
continuous count of days that is known as the Julian day number. The day number
is extended by the addition of the time of day, expressed as a decimal fraction of
day, to give the Julian date (JD). It is important to note, however, that the integral
values of the Julian date refer to the instants of Greenwich mean noon (since the
system was introduced when the astronomical day began at noon rather than at
midnight); correspondingly, the Julian date for 0h UT always ends in ".5". This,
and the fact that current Julian dates require seven digits to express the integral
parts, make the Julian date system inconvenient for some purposes. The value of
JD minus 2400000.5 is sometimes used for current dates; it is known as the modified
Julian date (MJD). The Julian date system may be used with TAI, TDT, or UTC,
so when the difference is significant, the particular timescale should be indicated.
1.253 Besselian and Julian Years For some astronomical purposes, such as the
specification of the epoch of a celestial coordinate system, it is convenient to measure
time in years and to identify an instant of time by giving the year and the decimal
fraction of the year to a few places. Such a system was introduced by Bessel, and is
still in use, but it has two disadvantages. First, the length of the year varies slowly,
and second, the instants at the beginning of the years (.0) do not correspond to
Julian dates, which are convenient for use in dynamical astronomy. The Besselian
system has been replaced by a new system in which 100 years is exactly 36525 days
(or 1 Julian century) and in which 1900.0 corresponds exactly to the epoch 1900
January 0.5, from which time interval was reckoned in the principal theories of the
motions of the Sun, Moon, and planets. The old and new systems are distinguished
when necessary by the use of the prefix letters "B" and "J." The standard epoch
that is now recommended for use for new star catalogs and theories of motion is
J2000.0 = 2000 January 1.5 = JD 2451545.0 TDB. (1.253-1)
1 .3 CELESTIAL AND TERRESTRIAL COORDINATES
1 .31 Coordinate Systems and Frames
In astronomy, it is necessary and convenient to represent the position of an object,
such as a star or a planet, in several different coordinate systems according to the
1 ( INTRODUCTION TO POSITIONAL ASTRONOMY 9
context in which the position is to be used. Each coordinate system corresponds to
a particular way of expressing the position of a point with respect to a coordinate
frame, such as a set of rectangular axes. Often the coordinate frames with respect
to which observations are made differ from the coordinate frames that are most
convenient for the comparison of observational data with theory. In general, each
position is represented by a set of three coordinates that specify the position of the
object with respect to a particular coordinate frame. In many cases, the distance of
the object is not known; so two coordinates are sufficient to represent the direction
to the object, although three direction cosines may be used.
In general, an object is moving with respect to the coordinate frame, and the
coordinate frame is moving and rotating in "space." Therefore, it is necessary to
specify the time to which the three spatial coordinates refer and the time for which
the coordinate frame is defined. These two times may be the same, but are often
different. Times may be expressed in terms of different timescales. Both spatial
coordinates and times may be expressed in a variety of units. As a further compli-
cation, positions may be of several kinds, according to whether or not allowances
have been made for aberration, diurnal rotation, refraction, and other factors that
affect the direction in which an object is observed. Ephemerides and catalogs that
represent high-precision positions and properties of astronomical objects at given
times must always be accompanied by a precise statement that specifies all of these
various factors, if the numbers are to be used properly to their full precision.
A coordinate frame is usually represented by three mutually perpendicular
(rectangular) axes, and is defined by specifying its origin, a fundamental reference
plane (the xy-plane), a direction in that plane (the *-axis), and the positive (or
north) side of the plane. Other fundamental planes or directions may be used; for
example, the direction of the z-axis (i.e., the normal to, or pole of, the ry-plane)
may be specified, or the jt-axis may be specified as the line of intersection, or node,
of two reference planes.
The vector r from the origin to the object may be represented by rectangular
coordinates {x,y,z)\ that is, by the projections of the distance r on the three axes.
It can also be represented by spherical coordinates, in which direction is usually
specified by the longitudinal angle (A) in the xy-reference plane and the latitudinal
angle (/?) from the reference plane (see Figure 1.31.1). Less frequently the polar
angle (90° -0), or complement of the latitudinal angle, is used; the prefix "co" may
be added to the name in such cases. The longitudinal and latitudinal angles are
given different names and symbols in different coordinate frames. If the longitudinal
angle is measured in the negative (or left-handed) sense with respect to the z-axis,
the prefix "co" may be used to indicate the complement to 360°.
The geometric transformations of coordinates between different coordinate sys-
tems, and the allowances for physical effects, may be carried out using the tech-
niques of spherical trigonometry or those of vector and matrix algebra. Generally,
matrices would be the preferred method, but some transformations are easier to
10
EXPLANATORY SUPPLEMENT
Z
X
Figure 1.31.1
Representation of the vector r in rectangular coordinates
explain using spherical trigonometry. The concept of the celestial sphere will be
used for purposes of explanation, with the arcs on the sphere representing angles
between directions. The center of the sphere can be located in many different places,
but in most cases the sphere will be used to illustrate a particular reference frame
with a specific origin. It must be remembered that when an object represented
on the sphere is changing radial distances from the center, these changes must be
incorporated into the mathematics of the computation.
1.311 Coordinate Designations Many different coordinate systems are used to
specify the positions of celestial objects. Each system depends on the choice of
coordinate frame and on the way of specifying coordinates with respect to the
frame. The term celestial coordinate frame is used here for a coordinate frame that
does not rotate with the Earth (or other planet), so that the coordinates of stars,
for example, change only slowly with time. (Frames that rotate with the Earth are
discussed in Section 1.33.)
The designations used to indicate the principal origins of celestial coordinate
frames are as follows:
(1) topocentric: viewed or measured from the surface of the Earth.
(2) geocentric: viewed or measured from the center of the Earth.
1 / INTRODUCTION TO POSITIONAL ASTRONOMY
11
(3) selenocentric: viewed or measured from the center of the Moon.
(4) planetocentric: viewed or measured from the center of a planet (with cor-
responding designations for individual planets).
(5) heliocentric: viewed or measured from the center of the Sun.
(6) barycentric: viewed or measured from the center of the mass of the solar
system (or of the Suti and a specified subset of planets).
The principal celestial reference planes through the appropriate origins are as
follows:
(1) horizon: the plane that is normal to the local vertical (or apparent direction
of gravity) and passes through the observer.
(2) local meridian: the plane that contains the local vertical and the direction
of the axis of rotation of the Earth.
(3) celestial equator: the plane that is normal to the axis of rotation of the
Earth and passes through the Earth's center.
(4) ecliptic: the mean plane (i.e., ignoring periodic perturbations) of the orbit
of the Earth around the Sun.
(5) planet's meridian: a plane that contains the axis of rotation of the planet
and passes through the observer.
(6) planet's equator: the plane that is normal to the axis of rotation of the
planet and passes through the planet's center.
(7) orbital plane: the plane of the orbit of a body around another (e.g., of a
planet around the Sun or barycenter).
(8) invariable plane or Laplacian plane: the plane that is normal to the axis of
angular momentum of a system and passes through its center.
(9) galactic equator: the plane through the central line of the local Galaxy
(Milky Way).
1 .32 Celestial Coordinate Systems
1.321 Equatorial and Ecliptic Frames The line of intersection of the mean plane
of the equator and the ecliptic defines the direction of the equinox (T). Using
this direction as the origin, the right ascension (a) is measured in the plane of
the equator and celestial (or ecliptic) longitude (A) is measured in the plane of the
ecliptic (see Figure 1.321.1). Right ascension, like hour angle, is usually expressed in
time measure from 0h to 24", and both right ascension and longitude are measured
in the positive (or right-handed) sense. The complement of right ascension with
respect to 24h is known as sidereal hour angle (SHA); in navigational publications
it is usually expressed in degrees. Declination is measured from the equatorial plane,
positive to the north, from 0° to 90°. Celestial latitude is measured from the ecliptic
plane, positive to the North, from 0° to 90°.
12
EXPLANATORY SUPPLEMENT
zi Pole of plane 0Xl,yi , equator
Pole of ecliptic
Equator of Zi
Ecliptic
Figure 1.321.1
Equatorial and ecliptic reference planes
The equinox is at the ascending node of the ecliptic on the equator; this is
the direction at which the Sun, in its annual apparent path around the Earth,
crosses the equator from south to north. It is also referred to as "the first point
of Aries," and it is the vernal (spring) equinox for the Northern Hemisphere. (See
Figure 1.321.1.) The inclination of the plane of the ecliptic to the plane of the
equator is known as the obliquity of the ecliptic (e).
The equator and ecliptic are moving because of the effects of perturbing forces
on the rotation and motion of the Earth. Hence, the equinox and obliquity change
as a function of time; so these celestial coordinate frames must be carefully defined
in such a way that they can be related to a standard frame that may be regarded
as being fixed in space.
1.322 Precession and Nutation The celestial pole, which is in the direction of the
axis of rotation of the Earth, is not fixed in space, but changes because of gravita-
tional forces, mainly from the Moon and Sun, that act on the nonspherical distri-
bution of matter within the Earth. The quasi-conical motion of the mean celestial
pole around the pole of the ecliptic is known as lunisolar precession. The much
smaller motion of the actual, or true, celestial pole around the mean pole is known
as nutation. The period of the precessionai motion is about 26,000 years; the prin-
cipal period of the nutation is 18.6 years. The motions of the celestial pole are
accompanied by corresponding motions of the celestial equator and of the equinox.
1 / INTRODUCTION TO POSITIONAL ASTRONOMY 1 3
The motion of the equinox due to precession gives rise to an almost steady
increase in the celestial longitudes of all stars at the rate of about 50" per year, but
the latitudes are hardly affected. The corresponding changes in right ascensions and
declinations depend on the position of the stars in the sky. The nutational motion
has an amplitude of about 9"; it is usually represented as the sum of two components,
one in longitude and one in obliquity. There is also a small rotation of the plane
of the ecliptic that gives a precession of the equinox of about 12™ per century and
a decrease of the obliquity of the ecliptic of about 47" per century; this is known
as planetary precession since it is due to the effects of planetary perturbations on
the motions of the Earth and Moon. These motions of the equatorial and ecliptic
coordinate frames can be modeled; however, some parameters have to be determined
from observations, since, for example, the internal structure of the Earth is not
known well enough.
In principle it is possible to obtain a standard celestial coordinate frame that is
fixed in space by using the frame that is appropriate to an arbitrarily chosen instant
of time, which is known as the standard epoch. The positions of the equator and
equinox for such a standard epoch cannot be observed directly, but must be specified
by adopting a catalog of the positions and motions of a set of stars or other celestial
objects that act as reference points in the sky. The standard epoch is now J2000.0,
although the epoch B1950.0 has been used for much of this century. There can be
different standard catalogs; for example, one is based on the positions of very distant,
quasars that have been measured by the techniques of Very-Long Baseline Radio
Interferometry (VLBI); another is the FK5 catalog of the positions and motions
of a much larger number of bright stars. The relative orientation of these different
frames is determined from observations. Standard expressions for precession and
nutation are used to connect positions for the standard frames (which are for the
standard epoch of J2000.0) with those for the mean and true frames of other dates.
A statement of the celestial coordinates of a star or other celestial object must
always specify precisely the coordinate frame to which they refer. It must also be
recognized that the coordinates that represent the position (direction) of a celestial
object depend on the position and motion of the observer as well as on the origin
and orientation of the coordinate frame to which they are referred.
1 .33 Terrestrial Coordinate Systems
For astronomical purposes it is necessary to be able to specify the position of an
observer, or other object, at or near the surface of the Earth in a geocentric coor-
dinate frame that rotates with the Earth and that may be related to the geocentric
celestial frame used for specifying the positions of the "fixed" stars. It is also nec-
essary to set up topocentric coordinate frames that may be regarded as fixed with
respect to the surface of the Earth and that may be used in the measurement of the
-|4 EXPLANATORY SUPPLEMENT
positions of celestial objects. The direction of the axis of rotation of the Earth pro-
vides an observable reference axis for both geocentric and topocentric frames, and
the direction of the vertical (or apparent direction of gravity) provides an additional
observable reference axis for topocentric frames.
1.331 Horizon Reference The lines of intersection of the plane of the meridian
with the planes of the horizon and equator define the directions from which azimuth
(A) and local hour angle (h) are measured. Azimuth is measured in the plane of
the horizon from the north, increasing in positive value toward the east. Local hour
angle is measured in units of time: 1 hour for each 15° positive to the west with
respect to the local meridian.
The latitudinal angles with respect to the horizon and equator are known as
altitude (a) and declination (6). Altitude is measured positively toward the zenith;
in astronomy the zenith distance (z = 90° — a) is more generally used. Declination
(6) is measured from the equator, positive toward the north pole of rotation. The
zenith distance of the north pole, which is the same as the codeclination of the local
vertical (or zenith), is equal to the geographic colatitude of the point of observation.
This relationship is the basis of the astronomical methods for the determination of
geographic latitude.
1.332 Shape of the Earth For some astronomical purposes the Earth may be re-
garded as a sphere, but often one must treat the Earth as an oblate ellipsoid of
revolution, or spheroid, and may need to take into account other departures from
sphericity. In particular, the height above mean sea level of a point on the Earth is
measured from an irregular surface that is known as the geoid (see Section 1.336).
Such heights satisfy the condition that water flows downhill under gravity. There are
several systems of terrestrial coordinates; so any set of precise coordinates should
be accompanied by a precise description of their basis.
The oblateness of the Earth is a consequence of the rotation of the Earth — the
axis of rotation coincides on average with the axis of principal moment of inertia,
which is usually referred to as the axis of figure. The axis of rotation and the axis
of figure do not coincide exactly, but the axis of rotation moves slowly in the Earth
around the axis of figure in a quasi-periodic motion with a maximum amplitude of
about 0['3 (or about 9m on the surface of the Earth). This is referred to as polar
motion. The equator of the Earth passes through the center of mass of the Earth.
1.333 Geocentric Coordinates Geocentric coordinates are coordinates that refer
to a coordinate frame whose origin is at the center of mass of the Earth and whose
fundamental planes are the equator and an (arbitrary) prime meridian through
"Greenwich" (see Figure 1.333.1). This frame defines a system of rectangular coor-
dinates (x,y,z) or of spherical coordinates, (longitude (A'), latitude (ip'% and radial
distance (p)). Longitude is measured positively to the east and usually designated
1 / INTRODUCTION TO POSITIONAL ASTRONOMY
15
z
North Pole
Equator
Figure 1.333.1
Geocentric and geodetic coordinates
from ± 180°. Longitude may be expressed in time units from 0h to 24h for some
applications. Latitude is measured from the equator (0°) to ± 90°, positively to the
north.
The Earth is not a rigid body, but is subject to secular and periodic changes
in shape and in the distribution of mass. The crust of the Earth may be consid-
ered to consist of plates that move slowly over the mantle and that are subject to
deformation. In these circumstances, the coordinate frame is realized by the adop-
tion of values for the positions and motions of a set of primary reference points
on the surface of the Earth. These values are chosen in an endeavor to satisfy cer-
tain conditions; for example, so that the prime meridian nominally passes through
Greenwich and so that there is no net rotation of the primary points \vith respect to
the frame. In practice, different techniques of observation and analysis give rise to
reference frames that differ from each other, and from the standard frame, in origin,
orientation, and scale. Moreover, the z-axis of the standard geocentric terrestrial
frame is not aligned with the Earth's present axis of figure, since it was chosen to
give continuity with the axis represented by the adopted coordinates of the original
set of monitoring observatories. This axis corresponds to the Conventional Inter-
national Origin of the coordinates of the pole, and so may be referred to as the
conventional pole.
-| g EXPLANATORY SUPPLEMENT
1.334 Geodetic Coordinates Geodetic coordinates are coordinates (longitude, lat-
itude, and height) that refer to an adopted spheroid whose equatorial radius and
flattening (or equivalent parameter) must be specified. Geodetic longitude (A) is
defined in the same way as geocentric longitude (A'), and these two coordinates
are equal to each other apart from any error in the adopted origin of geodetic
longitude. Geodetic latitude (tp) is equal to the inclination to the equatorial plane
of the normal to the spheroid (see Figure 1.333.1). Geodetic latitude may differ
from geocentric latitude (<p') by up to 10' in mid-latitudes. Geodetic height (h) is
the distance above the spheroid measured along the normal to the spheroid. (See
Section 4.21.)
Geodetic coordinates may be formally referred to an internationally adopted
spheroid whose origin and fundamental planes are those specified for geocentric
coordinates. Generally, however, they are referred to a spheroid that has been fitted
to the geoid over a particular region. The center of such a spheroid may differ
significantly from the center of mass of the Earth, and the fundamental planes
may not be parallel to those of the international spheroid. The orientation of the
reference frame of a regional spheroid is specified by the adopted coordinates of
the origin point, which is usually a geodetic reference station near the center of
the region concerned. Geodetic coordinates may also be referred to globally defined
spheroids whose origins and fundamental planes may be found to differ from the
ideal specification. Usually the term datum is used in geodesy for regional or global
reference frames such as the ones given in Section 4.24 for the international spheroid
and also for certain global and regional spheroids. In certain cases up-to-date values
of these parameters are given in The Astronomical Almanac.
1.335 Geographic Coordinates The term geographic coordinates is used for terres-
trial longitude and latitude when these are determined by astronomical observations
with respect to the celestial pole and the local meridian through the local vertical;
and for height above the geoid, commonly referred to as height above mean sea
level. The terms astronomical coordinates and terrestrial coordinates are also used
for such coordinates.
The geographic longitude of a point may be defined as the angle between the
plane of the astronomical meridian through the point and the plane of the prime
meridian through Greenwich. The astronomical meridian is the plane that contains
the direction of the local vertical and the direction of the line through the point that
is parallel to the axis of rotation of the Earth. The local vertical is affected by local
gravity anomalies and by the varying gravity fields of the Sun, Moon, and oceans.
Therefore, the astronomical meridian is not precisely the same as the geodetic
meridian that passes through the point and the actual axis of figure through the
center of the Earth. Consequently, the geographic and geodetic longitudes of a point
differ slightly. Geographic longitude is equal to the difference between (observed)
local sidereal time and Greenwich sidereal time at any instant.
1 / INTRODUCTION TO POSITIONAL ASTRONOMY 1 J
Geographic latitude is also defined in terms of the local vertical and the axis
of rotation, and so is not equal to the geodetic latitude. The inclination of the
local vertical to the normal to the reference spheroid is known as the deflection
of the vertical. The corresponding difference between the geographic and geodetic
coordinates may be as much as 1' (that is up to 4s in time- measure). The motion of
the axis of rotation within the Earth causes the values of geographic longitude and
latitude to vary by up to about 0'.'3 over a period of a few years. The daily tidal
variations in the direction of the vertical may reach about 0"01.
1.336 Height and Geoid The height (or altitude) of a point on (or near) the sur-
face of the Earth depends on the reference surface to which it is referred. Geode-
tic heights are measured with respect to a specified geodetic datum (or reference
spheroid), but the more commonly used geographic heights are measured with re-
spect to a surface that is known as the geoid. The geoid is a particular equipotential
surface in the gravity field of the Earth and is commonly referred to as the mean-
sea-level surface. The geoid is an irregular surface that may lie above or below
the international-reference spheroid by as much as 100 m; this difference is known
as the "undulation of the geoid" and should be taken into account in the precise
derivation of geocentric coordinates when the height above mean sea level is given.
For very precise studies, it is also necessary to take into account the variations in
height due to Earth tides and ocean loading (see Section 4.3).
1.337 Topocentric Coordinate Systems There are two main topocentric coordi-
nate frames. One is based on the direction of the local vertical, which defines the
plane of the horizon; the other is based on the direction of the Earth's axis of ro-
tation, which defines the plane of the celestial equator. The local meridian, which
contains the direction of the local vertical and the direction of the axis of rotation,
is common to both frames.
The angular coordinates in the topocentric coordinate frame of the horizon
and local meridian are known as azimuth and altitude. Azimuth (A) is measured
from 0° to 360° from north in the direction of east; that is, in the negative sense
with respect to the direction of the zenith. (In some circumstances azimuth may be
measured from the south, and so the convention that is used should be stated to
avoid ambiguity.) Altitude (a) is measured positively from the horizon toward the
zenith; however, in astronomy, the zenith distance (z = 90° — a) is more commonly
used. Both the altitude and the azimuth of any celestial object change rapidly as
the Earth rotates, and therefore, for many purposes it is more convenient to use the
topocentric frame of the celestial equator and the local meridian. The direction of
the north celestial pole is parallel to the Earth's axis of rotation, and the altitude
of the pole is equal to the geographic latitude of the point of observation. This is
illustrated in Figure 1.337.1, which shows the Earth as a sphere.
18
EXPLANATORY SUPPLEMENT
Plane of Earth's equator
Figure 1.337.1
Relation between geographic latitude and the altitude of the celestial pole
In this topocentric equatorial frame, the angular coordinates are known as local
hour angle (LHA, or h) and declination (6). LHA is measured from 0° to 360° , or
from 0h to 24h, from south in the direction of west (i.e., in the negative sense with
respect to the direction of the north celestial pole), and the LHA of a celestial object
increases by about 15°, or lh, for every hour of time. Declination is measured from
the equator to the celestial pole, positive toward the north. Codeclination, measured
from the pole toward the equator, is sometimes used instead of declination. The
declination of a celestial object is not changed by the diurnal rotation of the Earth.
1 ,34 The Rotation of the Earth
The connection between a local coordinate frame (defined with respect to the Earth
at the point of observation) and a celestial coordinate frame is complicated by
three aspects of the rotation of the Earth: the precession and nutation of the axis
of rotation (Section 1,322), the motion of the pole of rotation within the Earth
(Section 1.341), and the variability of the rate of rotation (Section 1.342).
1.341 Polar Motion Measurements of the zenith distances of the celestial pole of
rotation at different places and times show that the Earth's axis of rotation is not
fixed within the Earth, but moves around the axis of maximum moment of inertia.
This phenomenon is known as polar motion. Since it cannot be predicted accurately
1 / INTRODUCTION TO POSITIONAL ASTRONOMY -| 9
in advance, the position of the pole is determined from regular observations as an
international service. In a celestial coordinate system, the axis of maximum moment
of inertia moves around the axis of rotation in a complex motion in which an annual
component and a 14-month (Chandler) component create beats. The amplitude of
the quasi-circular motion varies between 0"05 and 0"25 in a six-year cycle.
1.342 Length of Day The variations in the Earth's rate of rotation are, at present,
unpredictable and are due to several different causes. There is a long-term slowing
down of the rate of rotation that appears to be due largely to the effect of tidal
friction. There are irregular changes with characteristic periods of a few years that
appear to be due to an interchange of angular momentum between the mantle and
the core of the Earth. There are also seasonal and short-period irregular changes
that appear to be due largely to exchanges of angular momentum between the
solid Earth and the atmosphere. The total effect on the orientation of the Earth in
space is determined from observations by an international monitoring service that
provides future predictions based on the observations.
These variations are usually represented by the corresponding changes in the
length of the (universal) day (with respect to a day of 86400 SI seconds) and
by the accumulated difference between UT and TAI (or ET before 1955). The
latter difference in time may be expressed as a corresponding difference in angular
orientation of the Earth. Table 1.342.1 shows the orders of magnitude of the different
variations.
The accumulated effect of the changes in the rate of rotation can be seen most
clearly in the geographic longitudes and times at which certain phenomena, such
as eclipses of the Sun, were observed in early history. The accumulated effect of
the changes in the rate of rotation has amounted to two hours over the last two-
thousand years. This was a source of much confusion when the rotation of the
Earth provided the standard timescale for all purposes. It is now recognized that
sidereal time and Universal Time are best regarded as measures of angles that serve
to specify the orientation of terrestrial coordinate systems with respect to celestial
Table 1.342.1
Variations in the Earth's Rate of Rotation
Characteristic
Change in
Difference
Difference
Time
Length of Day
in Time
in Orientation
a few days
0. 1 ms
1 ms
o:'oi
100 days
1 ms
0.2s
3"
a few years
3ms
10s
3'
20 EXPLANATORY SUPPLEMENT
coordinate systems. They must be determined from observations, although man-
made clocks can be used for interpolation and extrapolation.
1 .35 The Connections between Terrestrial and Celestial Coordinates
In studies of the distribution and motion of celestial bodies it is necessary to have
procedures for the transformation, or reduction, of "observed" topocentric coor-
dinates, which are referred to a terrestrial frame that rotates with the Earth, to
(or from) "reduced" barycentric coordinates, which are referred to a celestial frame
that is fixed in space. The generally adopted procedure uses an intermediate system
of geocentric coordinates, which are referred to a celestial equatorial frame of date
that is defined by the current directions of the Earth's axis of rotation and of the
equinox. The principal stages of the full reduction are as follows:
Stage 1: Rotation of the terrestrial frame from the horizon to the equator and
from the local meridian to the prime meridian.
Stage 2: Translation of the origin of the terrestrial frame from the observer to the
geocenter — a change from UTC to TDT may be required.
Stage 3: Rotation of the terrestrial frame for polar motion; i.e., from the conven-
tional pole to the true celestial pole (or celestial ephemeris pole).
Stage 4: Rotation of the terrestrial frame around the true pole from the Green-
wich meridian to the celestial frame whose prime meridian passes through the true
equinox. (The angle of rotation varies rapidly as the Earth rotates, and is known
as the Greenwich apparent sidereal time.)
Stage 5: Rotation of the celestial frame for nutation from the true pole and true
equinox to the mean pole and mean equinox of date.
Stage 6: Rotation of the celestial frame for precession from the mean equator and
equinox of date to the standard (mean) equator and equinox of J2000.0.
Stage 7: Translation of the origin from the geocenter to the barycenter of the Solar
System — a change from TDT to TDB may be required.
Stage 1 requires a knowledge of the geographic longitude, latitude, and height
of the place of observation. If appropriate, the effect of the tides on the deflection
of the vertical and on the height may have to be taken into account.
Stage 2 requires a knowledge of the geodetic longitude, latitude, and height for
reducing the coordinates of very close objects. For distant objects, such as stars,
1 / INTRODUCTION TO POSITIONAL ASTRONOMY 21
the translations of the origin may be treated as small rotations corresponding to
small parallactic shifts in direction.
The standard values of the coordinates of the pole that are required in Stage 3
are published in arrears; "quick-look" and predicted values are also made available.
Indicative values are tabulated at a wide interval in The Astronomical Almanac (p.
K10).
For Stage 4, the required value of the Greenwich sidereal angle at the time
of observation may be calculated in three steps from the UTC of the instant of
observation. The first step is to apply the difference between UT and UTC values
to the value of UTC. Standard values are published in arrears, but a quick-look
or predicted value may be of sufficient accuracy. The value of UT is then used
to compute the values of the Greenwich mean sidereal time from a polynomial
expression. Finally, the equation of the equinoxes is applied to give the Greenwich
apparent sidereal time, which is converted to angular measure to give the required
Greenwich sidereal angle.
The values of the nutation in longitude and obliquity that are required in Stage
5 may be computed from harmonic series that are based on an adopted theory
of nutation. In some circumstances it may be appropriate to apply further small
corrections obtained from the analysis of recent VLBI observations.
The values of the precession that are required in Stage 6 may be computed
from adopted expressions for the angular quantities concerned. Stages 5 and 6 may
be combined in a single rotation matrix; daily values of the matrix are published
in The Astronomical Almanac (pp. B45-B59).
The rectangular coordinates of the geocenter with respect to the barycenter may
be obtained from published ephemerides, such as that printed in The Astronomical
Almanac (pp. B44-B58), which gives both position and velocity.
1 .36 Effects of the Position and Motion of the Object and Observer
1.361 Parallax The direction in which an object is seen depends on the position
of the observer as well as on the position of the object. The change in the direction
of a celestial object caused by a change in the position of the observer is known as
a parallactic shift. The amount of change in direction corresponding to a standard
linear change in the position of the observer, across the line of sight, is known as the
parallax of the object — it is an inverse measure of the distance of the object from
the observer (see Figure 1.361.1). For objects within the solar system the standard
linear change is the equatorial radius of the Earth. For the stars and other such
distant objects, however, the standard linear change is the astronomical unit of
distance (au) or, in effect, the mean radius of the Earth's orbit around the Sun.
The basis of the parallax can be deduced immediately from a knowledge of the
type of object. The parallax of an object in the solar system is usually referred to
as a horizontal parallax, since the actual parallactic shift takes this value when the
22
EXPLANATORY SUPPLEMENT
Earth
Earth
orbit
Earth
Figure 1.361.1
Parallax of an objecl
object is seen on the horizon (from a point on the Earth's equator). The parallax
of a star is often referred to as an annual parallax, since the parallactic shift varies
with a period of one year. By analogy, a horizontal parallax may be referred to as
a diurnal parallax; the oblateness of the Earth must be taken into account in the
precise computation of the parallactic shift of the Moon.
Distances within the solar system are conveniently and precisely expressed in
astronomical units and may be expressed in meters using a conversion factor based
on radar measurements to planets. Prior to 1960, this conversion factor was based
on determinations of the solar parallax, i.e., of the equatorial horizontal parallax of
the Sun when at a distance of 1 AU.
The catalog positions of stars and other objects outside the solar system are
usually referred to the barycenter of the solar system. The correction for annual
parallax may be applied with sufficient accuracy for many purposes by assuming
that the orbit of the Earth is circular. For full precision, however, it is usual to
express the correction in terms of the barycentric rectangular coordinates of the
Earth that are tabulated in The Astronomical Almanac. These coordinates are also
used to reduce the times of observation of such phenomena as pulsar emissions to
a common origin at the barycenter of the solar system.
1.362 Proper Motion and Radial Velocity The mot ion of an object outside the so-
lar system is usually expressed in terms of its proper motion and radial velocity.
Proper motion is expressed in terms of the secular rates of change in the right as-
cension and declination with respect to a standard coordinate frame. It is necessary
to know the distance (or parallax) of the object before the motion transverse to the
line of sight can be expressed in linear measure. Radial velocity is the name given to
the motion of the object along the line of sight and is derived from observations of
the Doppler shift in the radiation from the object. It is usually expressed in km/s.
An observational value must be corrected for the rotational motion of the point of
1 / INTRODUCTION TO POSITIONAL ASTRONOMY
23
observation and the orbital motion of the Earth around the barycenter. When the
parallax is known, the proper motion and radial velocity can be combined to form
a space-motion vector that represents the motion of the object with respect to the
standard reference frame. This space-motion vector should be used when comput-
ing apparent positions for any date and transforming positions from one standard
epoch to another.
1.363 Aberration Because the velocity of light is finite, the direction in which a
moving celestial object is seen by a moving observer (its apparent position) is not
the same as the direction of the straight line between the observer and the object
(its geometric position) at the same instant. This displacement of the apparent
position from the geometric position may be attributed in part to the motion of
the object and in part to the orbital and rotational motion of the observer, both
of these motions being referred to a standard coordinate frame. The former part,
which is independent of the motion of the observer, is referred to as the correction
for light-time; the latter part, which is independent of the motion or distance of the
object, is referred to as stellar aberration. (For stars the correction for light-time
is ignored.) The sum of the two parts is called planetary aberration because it is
applicable to planets and other members of the solar system. The term aberration
is used in this supplement to include the effects of the motions of both object and
observer, but it is sometimes used elsewhere for the effect due to the motion of the
observer alone. For precise astrometry, stellar aberration must be calculated from
the formulas of the special theory of relativity, but for most classical purposes the
theory depending on only the first power in vie, where v is the velocity of the
observer and c is the velocity of light, is sufficient.
The motion of an observer on the Earth is the resultant of the Earth's diur-
nal rotation and the Earth's annual orbital motion around the barycenter of the
solar system. Stellar aberration is therefore regarded as the sum of diurnal aberra-
tion and annual aberration, for which the maximum effects are about 1'.'3 and 20",
respectively.
Annual aberration affects all objects in the same part of the sky in a similar
way, and so in photographic observations of stars it does not obscure the much
smaller effect of annual parallax, which affects the apparent positions of only the
nearby stars.
In determining precisely the differences between the celestial coordinates of
a solar system object and a fixed star, it is necessary to take into account the
differential aberration between the object and the star. In constructing an ephemeris
of a moving object that is to be compared directly with the catalog positions of
stars at some epoch, it is necessary to take into account the correction for light-time
but not stellar aberration; such an ephemeris is known as an astrometric ephemeris.
An ephemeris in which aberration is fully taken into account along with precession,
nutation, and light-bending is known as an apparent ephemeris.
24 EXPLANATORY SUPPLEMENT
1.364 Deflection of Light In precise computations it is necessary to take into ac-
count the effect of the gravitational field of the Sun (or other massive object) on
the velocity of propagation of light. The effect causes a bending of the path, so that
the apparent angular distance from the Sun is greater, and an increase in the travel
time, so that pulsed radiation is retarded. This effect resembles refraction in the
Earth's atmosphere, although the latter effect is usually much greater.
1.365 Refraction Refraction in the Earth's atmosphere affects most directional
observations. It depends largely on the altitude of the object above the horizon,
and its precise value depends on the atmospheric conditions at the time. Since the
effect depends on the position of the object with respect to the observer, it cannot
be included in ephemerides. Nominal allowances are made for refraction in the
computations of the times of rising and setting phenomena, for which the effects of
refraction are greatest. Refraction increases the apparent altitude of a celestial body
and increases the travel times of laser pulses. Most analytical theories and tables
of refraction assume that the surfaces of equal refractive index in the atmosphere
are concentric and that it is sufficient to know the pressure, temperature, and
relative humidity at the point of observation. Significant variation from such models
can occur, so that the angular refraction may not be exactly zero in the zenith.
Measurements of azimuth may also be affected by refraction in the horizontal plane.
The refraction of radio waves depends largely on the properties of the iono-
sphere and is much more variable than refraction for light waves. Multiple-frequency
measurements may be used to determine the refraction correction for radio waves.
1.4 ORBITAL MOTIONS
1 .41 Motion in Two-body Systems
1.411 Laws of Motion Over a short period of time the motion of a planet around
the Sun, or of a satellite around a planet, can usually be treated as if the two
objects are isolated in space and moving around each other in accordance with
Newton's laws of motion and gravitation (i.e., Keplerian or two-body motion). The
smaller, or secondary, body moves around the larger, or primary, body in an orbit
that lies in a plane passing through the primary. The orbit is in the shape of an
ellipse with the primary at a focus. Correspondingly, the larger body moves in a
smaller orbit around the center of mass of the pair. The orbits of the major planets
and most satellites are almost circular, whereas the orbit of Comet Halley, for
example, is extremely elongated. The shape of an ellipse is usually characterized
by its eccentricity, e. A circle is a limiting case as e approaches zero, and e tends
to unity with increasing elongation. Comets may also move in parabolic (e - 1)
or hyperbolic (e > 1 ) orbits. The geometric properties of these conic sections are
summarized in Figure 1.411.1.
26
EXPLANATORY SUPPLEMENT
Kepler discovered the elliptical character of the orbit of Mars, and found that
the speed of motion of Mars was such that the line joining Mars to the Sun swept
out equal areas in equal times. He also found that the squares of the periods of
revolution (P) of the planets are proportional to the cubes of the lengths of their
semi-major axes (a). Newton was the first to demonstrate that these properties
were consistent with the existence of a gravitational force of attraction between the
Sun and the planet. This attraction varies inversely as the square of the distance
(r) between the two bodies and directly as the product of their masses (M and m).
Newton defined the mass of a body as a measure of the amount of matter in the
body, and the momentum of a body as the product of its mass and its velocity.
His second law of motion states that the rate of change of momentum of a body is
proportional to the force acting on it and is in the same direction.
If the Sun and a planet are regarded as forming an isolated system, then the
system's center of mass can be regarded as being at rest or moving at a uniform
velocity in space. This point can be treated as the origin of a set of nonrotating
coordinate axes, which are then used to form an inertial coordinate frame to which
Newton's laws of motion and gravitation apply. One can then write the equations
of motion of the Sun and of the planet with respect to the frame. The equation for
the motion of the planet relative to the Sun (i.e., treating the Sun as the origin of
a coordinate frame whose axes remain parallel to those of the inertial frame) may
then be derived; it takes the simple form
G(M + m)r
r= j3 , (1.411-1)
where r and f are the position and acceleration of the planet with respect to the
Sun expressed as vectors, with Newton's dot notation to denote differentiation
with respect to time, and where G is the constant of gravitation. If the vector r has
components x, y, z with respect to the coordinate axes, then Equation 1.411-1 is
equivalent to three separate equations of the form
G(M + m)x
x= . (1.411-2)
Many of the basic properties of motion in an ellipse may be derived by consider-
ing the case when the orbit lies in the xy-plane and expressing x and y in polar
coordinates r and 9. The principal results are given in Figure 1.411.1a.
It is convenient to use an astronomical system of units in which the unit of
mass is equal to the mass of the Sun, and the unit of time is one day. When this is
done, Equation 1.411-1 is written as:
k2{ \ +m)r
r= 3 , (1.411-3)
1 / INTRODUCTION TO POSITIONAL ASTRONOMY
27
Figure 1.412.1
Angular orbital elements
where k is known as the Gaussian constant of gravitation. It is now the practice to
treat the value of k as a fixed constant that serves to define the astronomical unit
of length (au) which is slightly less than the semi-major axis of the orbit of the
Earth around the Sun (see Chapter 15).
1.412 Orbital Parameters Positions of a body moving in an elliptical orbit can
be computed from a set of six orbital parameters (or elements) that specify the
size, shape, and orientation of the orbit in space and the position of the body at a
particular instant in time known as the epoch. Commonly used parameters are:
a = the semi-major axis of the ellipse;
e = the eccentricity of the ellipse;
i = the inclination of the orbital plane to the reference plane;
£2 - the longitude of the node (N) of the orbital plane on the reference plane;
w = the argument of the pericenter (P) (angle from N to P);
T = the epoch at which the body is at the pericenter.
The longitude of the pericenter, <D, which is the sum of £2 and uj, may be used
instead of uj. And the value of t he mean anomaly, M, or mean longitude (L = Gj+M)
at an arbitrary epoch may be used instead of T. The reference plane is usually the
ecliptic, and the epoch for which the reference plane is specified must be stated.
These angular parameters are shown in Figure 1.412.1. The mean daily motion of
2g EXPLANATORY SUPPLEMENT
a planet (n) in its orbit may be determined from
n=Y = —^- (1.412-1)
where P is the period in mean solar days and n is in radians per day.
The elements may be given as mean elements where the values are a type of
average value over some period of time. The elements may be given as osculating
elements, which are the instantaneous values of the elements or the elements for
the elliptical motion that the body would follow if perturbing effects were removed.
Since the solar system bodies are in well-behaved orbits, this set of elements may
be used. In cases where the eccentricity or inclination may be zero, a universal set
of parameters should be used instead.
The orbital parameters can be used to compute the velocity as well as the
position of the body in its orbit. Conversely, if the position and velocity of a body
are known at a given instant, it is possible to calculate the parameters of the
elliptical orbit. When a new minor planet is discovered, each observation gives two
coordinates (usually, right ascension and declination). At least three observations,
preferably more, are required to determine the parameters of an elliptical orbit that
would pass through the observed positions. Gauss' and Laplace's methods can be
used for this procedure (Herget, 1948; Danby, 1989).
1 .42 Types of Perturbations
It is only for applications where reduced accuracy is required that two-body motion
is adequate for the representation of the motion of a planet or satellite. For other
applications a number of perturbation effects must be included:
(a) The attractions of other planets or satellites must be included in all cases.
(b) The nonsphericity of the distribution of mass in a primary body must be
included in calculating the motions of satellites.
(c) The tidal forces that arise when the gravitational attraction departs from
simple inverse square form must be included in the motions of satellites.
(d) The nongravitational forces, or reactive forces, must be included when com-
puting the orbit of a comet.
(e) The effect of atmospheric drag and solar wind must be included in the
computation of orbits of artificial satellites.
(f ) Radiation pressure must be included in the motion of artificial satellites and
in the motion of particles in the rings around planets.
1.421 Ephemerides For precise calculations of the motion of the planets and the
Moon, the equations of motion must include the relativistic terms. This is normally
done using the parameterized post-Newtonian (PPN) version of the equations. The
1 / INTRODUCTION TO POSITIONAL ASTRONOMY
29
independent variable, time, in the PPN equation for the solar system is the Barycen-
tric Dynamical Time. When converting the barycentric positions of the planets to
geocentric positions, one must convert the time according to the relativistic metric
used in the computation. Similarly, the optical observations being compared to the
ephemerides must be corrected for gravitational light bending, and radar observa-
tions must be corrected for the appropriate time delay involved. With the exception
of the Moon, the natural satellite ephemerides do not require the inclusion of rela-
tivistic effects to the accuracies currently being used.
The perturbation effects may be computed either by numerical integration of
the motion of the body or by the use of a general theory that provides an expression
for the motion of the body as a function of time. Currently, it is not possible to
achieve the accuracy of observations by means of a general-theory approach for
the motion of the planets and the Moon. On the other hand, for rapidly moving
satellites, such as the Galilean satellites, it is not possible to numerically integrate
those bodies over an extended period of time and retain the necessary accuracy
required for observations.
1.422 Numerical Integration Numerical integration is the most accurate method
of calculating the motion of bodies in the solar system, particularly the Moon and
planets. There are many methods of numerical integration. Textbooks on celestial
mechanics describe such methods. Usually, Cowell's method is used with an equa-
tion that includes the PPN relativistic formulation. The initial conditions for the
numerical integration are adjusted to fit available observational data and a sec-
ond numerical integration is performed based on that correction. This process is
repeated until the numerical integration represents the observational data to the
required accuracy.
1.423 Analytical Theories As an alternative to numerical integration, an analyt-
ical theory, usually in the form of a Fourier series or Chebyshev series, can be
developed to represent the motion of the bodies. Historically, this was the method
of computation for the planets and the Moon, and the technique is represented by
Newcomb's theories of the Sun, Mercury, and Venus; Hill's theory of Jupiter and
Saturn; and Brown's lunar theory. Today this method is used only for planetary
satellites, where sufficient accuracy can be achieved by this method. These theories
permit determination of the positions of the bodies for any time within the period
of validity for the theory. Theories of the planets and the Moon have been developed
by Bretagnon, Simon, Chapront, and Chapront-Touze, but they do not achieve the
accuracies of numerical integration.
1.424 Observational Data The computation of accurate ephemerides of the Sun,
Moon, and planets depends upon the availability of accurate observations of their
positions at precisely known times. Optical observations of the Sun, Moon, and
30 EXPLANATORY SUPPLEMENT
planets have been made using meridian, or transit circles; with a person's eye; or
with electronic detectors. Observations of Mercury, Venus, and Mars can be made by
radar ranging. Observations of the Moon have been made using laser ranging since
1969. Planetary radar-ranging and Lunar Laser-Ranging provide much more precise
measurements than transit-circle observations. Radio observations of spacecraft on
a planet, in orbit around a planet, or encountering a planet also provide very precise
positional data.
Observations of the satellites can be made by photographic or charge-coupled
devices (CCD) techniques, and by recording the times of mutual phenomena (eclipses,
transits, or occupations) of these bodies. Accurate observational data go back to
only about 1830; so computations of the motions of bodies in our solar system are
limited by the 160 years of observational history.
1.425 Differential Corrections The process of correcting a numerical integration
or an analytical theory to give a better fit to the observational data is referred to as
the differential correction process. In this process, an equation of condition can be
written for each observation. This equation provides the relationship between cor-
rections to the initial adopted conditions of the motion and the difference between
the computed and observed position of the body. These equations of condition are
combined into normal equations, which are then solved to determine the improved
initial conditions for the motions. This process is described in textbooks on celestial
mechanics.
1.426 Representative Ephemerides It is often advantageous to use computers to
compute the positions of the bodies from formulas rather than to read ephemerides
that were based on either numerical integrations or analytical theories. A represen-
tative ephemeris can be prepared by evaluating a limited expression that represents
the motion of the body to a specified level of accuracy over a limited time period.
These expressions, which are usually power series or Chebyshev polynomials, pro-
vide the most efficient means of generating the ephemerides with computers.
1.43 Perturbations by and on Extended Bodies
1.431 Gravity Field of Extended Bodies Newton's law of gravitation formally ap-
plies to idealized point masses, but it can be shown that the gravitational force due
to a spherically symmetric body is the same as if all the mass were concentrated at
the center. In practice, astronomical bodies are rarely spherically symmetric, and
most are rotating at such a rate as to cause the bodies to be clearly oblate. In ad-
dition, surfaces of equal density are not spherical, and the gravitational attraction
on an external point mass is not normally in the radial direction. Other departures
from purely radial variation in density occur; so the external gravity field of an
extended body, such as a, planet, is represented by a spherical harmonic series. The
1 / INTRODUCTION TO POSITIONAL ASTRONOMY
31
coefficients of such a series may be determined by analyzing the departures of the
motion of a satellite from a fixed ellipse. The effects are most noticeable for close
satellites, and the effects of the higher-order terms decrease very rapidly with dis-
tance. The most noticeable effect is that of the second-order zonal harmonic, J2
which causes the pole of a satellite orbit to precess around the polar axis of the
planet, though the inclination to the equator remains unaltered.. The direction of
the pericenter also precesses around the orbit. (See Section 4.31.)
The equations of motion of a satellite may be referred to a coordinate system
that is fixed relative to the planet if the appropriate Coriolis terms are included to
take account of the rotation of the planet. The expression for the potential function
may also be modified in a similar way— the arbitrary constant in the potential
function is usually chosen so that the surface of zero potential corresponds to mean
sea level (or the geoid) in the case of the Earth. This is the surface from which
height is measured for mapping purposes.
1.432 Forces on Extended Bodies The gravitational forces on an extended body
may be resolved into a force and a couple. The force acts through the center of
mass and causes an acceleration of the body in its orbit. The couple changes the
rotational motion of the body. A simple example is that of a two-body system,
such as the Sun and a planet, in which one body is an oblate spheroid that is
rotating about its principal axis. The torque, which may be considered to act on
the equatorial bulge, has a gyroscopic effect, and causes the principal axis to precess
around the normal to the orbit plane at a constant inclination. In more complex
systems, a nutational motion is superimposed on the main precessional motion, and
an irregular body may "tumble" as it moves around its orbit.
The forces on an extended body will also cause a tidal distortion even if it is a
solid, but slightly elastic, body. The effect is similar to that of the ocean tides on
the Earth. For a simple model there are two "high tides" and two "low tides" that
remain stationary with respect to the attracting body. The line of the high tides is
slightly displaced from the line joining the centers of the two bodies. This causes
a net torque that tends to change the rate of rotation of the body. On the other
hand, the existence of the tidal distortion changes the attraction on the other body
and gives a small component at right angles to the line of centers, thus causing the
orbit to expand or contract, depending on the circumstances. This effect is referred
to as tidal friction.
The principal elastic properties of a spherical body may be represented in terms
of parameters known as Love numbers. Resonance effects may occur if the tidal
forces contain a periodic component that matches one of the natural frequencies
of vibration of the body. The dissipation of the tidal energy may be large enough
to heat the body significantly; this is believed to be the case for Io of the Galilean
satellites of Jupiter.
32 EXPLANATORY SUPPLEMENT
The forces that give rise to the nutation of the rotation and the tidal distortion
of the shape of a body are essentially the same, although in the past t hey have often
been expressed in different forms, and with different notations. These forces can be
derived from a potential function that allows the various periodic terms in the
nutation and in the tides to be identified and computed. The accurate computation
of the Earth's nutation requires the adoption of a model of its elastic properties
and also of the viscous properties and shape of the fluid core.
1 .5 ASTRONOMICAL PHENOMENA
The various motions involved in the solar system, including the rotation of the
Earth, the orbital motion of the Earth and planets, and the motion of the Moon
around the Earth, cause a number of important and/or interesting astronomical
phenomena. Some of these phenomena are events that can be timed exactly, some
are matters of arbitrary definitions, and others are not subject to either exact timing
or exact definitions.
1.51 Rising, Setting, and Twilight
The rising and setting times of an object are given with respect to sea level, a
clear horizon, and refraction based on normal meteorological conditions. Rising
and setting times are strictly local phenomena depending upon the longitude and
latitude of the place on the Earth's surface. For the Sun and Moon, the times of
rising and setting refer to the instants when the upper limb of the body appears on
the apparent horizon.
The actual times of rising and setting may differ considerably from the tabu-
lated values, especially near extreme conditions when the altitude is changing very
slowly. Thus, the computation of the time of moonrise and moonset for very north-
ern or southern latitudes can be subject to large uncertainties. Also, the use of
imprecise methods can lead to extremely large errors. In addition, the illumination
at the beginning or the end of twilight can vary greatly because of meteorological
conditions and the physical surroundings. Precise times have little real significance
except in special circumstances, such as navigation at sea.
1.511 Times of Sunrise and Sunset The calculated times of sunrise and sunset are
based on the Sun having a geocentric zenith distance of 90°50' (34' is allowed for
horizontal refraction using a standard atmospheric model, and 16' for the semi-
diameter). Corrections are necessary if the meteorological conditions differ signif-
icantly from the standard model or to allow for the height of the observer or the
elevation of the actual horizon. The times of rising and setting and associated phe-
nomena change rapidly from day to day in polar regions or may not occur for long
1 / INTRODUCTION TO POSITIONAL ASTRONOMY 33
periods. When the Sun is continuously very near the horizon, accurate times are
difficult to tabulate.
1.512 Twilight Conditions Twilight is caused when sunlight is scattered by the
upper layers of the Earth's atmosphere. Twilight takes place before sunrise or after
sunset. There are three distinct types of twilight. Civil twilight ends in the evening
or begins in the morning when the center of the Sun reaches a zenith distance of
96°. Nautical twilight ends or begins when the Sun reaches a zenith distance of
102°. Astronomical twilight ends or begins when the Sun reaches a zenith distance
of 108°. In good weather conditions and in the absence of other illumination, the
indirect illumination of the Sun at the beginning or end of civil twilight allows
the brightest stars to be visible and the sea horizon to be clearly defined. For the
beginning or end of nautical twilight, the sea horizon is generally not visible, and
it is too dark for observations of altitudes with reference to the horizon. At the
beginning or end of astronomical twilight, the indirect illumination from the Sun
is less than the contribution from starlight and is of the same order as that from
the aurora, airglow, zodiacal light, and the gegenschein. In all cases the actual
brightness of the sky depends on direction as well as meteorological conditions.
1 .52 Meridian Transit
Once each day at a given location the Earth's rotation causes a celestial body to
cross the meridian plane defined by the north and south poles and the zenith. When
this happens, the object will have its maximum altitude with respect to the horizon.
Generally, observations with a telescope are best made when the object has an hour
angle of less than two hours.
For objects near the pole that can be observed as circumpolar objects from a
given location, there can be a second meridian transit when the object is below
the pole (lower transit). Observations when the object is above the pole are called
upper transit. These two observations permit an independent determination of the
location of the pole.
1 .53 Conjunction, Opposition, and Elongation
The times of conjunction and opposition of the planets are those at which the
difference between the apparent geocentric longitudes of the planet and the Sun
are 0° and 180°, respectively. Conjunctions for the outer planets take place when
the planets are in the direction of the Sun from the Earth. Oppositions occur when
the planets are in the opposite direction from the Sun with respect to the Earth. For
Mercury and Venus, superior conjunctions take place when the planets are more
distant from the Earth than the Sun, and inferior conjunctions occur when the
planets are closer to the Earth than the Sun.
34 EXPLANATORY SUPPLEMENT
If heliocentric geometric longitudes are used in place of geocentric longitudes,
a small correction must be applied to correct for the effect of light-time. This
correction can be a maximum of 10 minutes for the conjunctions of Venus and
Mars. Due to the eccentricities and inclinations of the orbits and the effect of
perturbations, the times for these phenomena may be different from those at which
the geocentric distance is a minimum at inferior conjunction of Mercury or Venus or
opposition of a superior planet, or at which the distance is a maximum at superior
conjunction of Mercury or Venus or conjunction of a superior planet.
The elongation {£) of a planet, measured eastward or westward from the Sun,
is calculated as the angle formed by the planet, the Earth, and the Sun. The elonga-
tions are measured from 0° to 180° east or west of the Sun and are tabulated to the
nearest degree. Because of the inclination of the planetary orbits, the elongations
do not necessarily pass through 0° or 180° as they change from east to west or west
to east.
1 .54 Eclipses, Occultations, and Transits
An eclipse takes place when one body passes into the shadow of another body. An
occultation takes place when a larger body passes in front of a smaller body so
that it cannot be seen by the observer. A solar eclipse takes place when the Moon
blocks the light of the Sun as seen by an observer on Earth, and so is really an
occultation. A transit takes place when a smaller body passes in front of a larger
body, e.g., when a satellite passes in front of a planet or when a planet passes in
front of the Sun. A shadow transit takes place when the shadow of a smaller body
passes in front of a larger body. This can occur when the shadows of satellites pass
in front of planets.
1.541 Transits Transits of Mercury and Venus across the Sun's disk can occur
only when both the Earth and the planet are simultaneously very close to the
same node of the planet's orbit on the ecliptic. Because of the near- constancy of
the longitude of the nodes and perihelia of the planets, the Earth will be near the
nodes on about the same dates each year, and the planets will be at the same point
in their orbits when they pass through the ecliptic. For the highly eccentric orbit
of Mercury, this means that the conditions and limits at the November transits are
very different from those at the May transits. During the twentieth century there
are fourteen transits of Mercury — ten in November and four in May, including
grazing transits in 1937 May and 1999 November. Future transits will take place
on 1993 November 6 and 1999 November 15. In the twenty-first century there will
be five transits in May and nine in November. The transits of Venus are quite rare.
Generally there are two transits of Venus within an eight-year period and then a
gap of over 100 years until the next pair of transits. The next transits of Venus will
be 2004 June 8 and 2012 June 6.
1 / INTRODUCTION TO POSITIONAL ASTRONOMY 35
1.542 Solar Eclipses Eclipses of the Sun or Moon occur when the centers of the
Sun, Earth, and Moon are nearly in a straight line. This condition can be fulfilled
only when conjunction or opposition occurs in the vicinity of a node of the lunar
orbit on the ecliptic. A total solar eclipse takes place when the disk of the Moon
appears to completely cover the disk of the Sun. Alternately, an annular solar eclipse
takes place when the apparent disk of the Moon is smaller than the apparent disk
of the Sun, so that the fight of the Sun is not completely blocked. A partial eclipse
of the Sun takes place when the location on the Earth is not in the path of the
umbral shadow cast by the Moon. The exterior tangents to the surfaces of the Sun
and Moon form the umbral shadow cone. The interior tangents form the penumbral
cone. The common axis of the cones is the axis of the shadow. The actual size of
the shadow will be rather small, leading to a narrow path and a short duration
for an eclipse. The path of the shadow on the surface of the Earth can be rather
complicated, due to the orbital motions of the Earth and Moon, and the rotation
of the Earth. Seven minutes of total solar eclipse is a rather long period for totality.
The actual length of time and width of the path varies for solar eclipses. Solar
eclipses come in certain patterns that repeat after a long interval called a Saros,
A Saros is approximately 223 lunations, which is approximately 19 passages of the
Sun through a node or approximately 6585 1/3 days (see Chapter 8).
1.543 Lunar Eclipses A lunar eclipse takes place when the Moon passes through
the shadow of the Earth. In this case, the times and circumstances are the same
for all parts of the Earth from which the Moon is visible. There are three types
of lunar eclipses: A penumbral eclipse, also called an appulse, takes place when the
Moon enters only the penumbra of the Earth. A partial eclipse occurs when the
Moon enters the umbra without being entirely immersed in it. A total eclipse takes
place when the Moon is entirely immersed within the umbra.
1 .55 Satellite Phenomena
The Galilean satellites of Jupiter undergo the phenomena of eclipses, occultations,
transits, and shadow transits. The times for beginning or ending of these phenomena
(i.e., the disappearance and reappearance for eclipses and occultations, and the
ingress and egress for transits and shadow transits) are given in The Astronomical
Almanac. When Jupiter is in opposition, the shadow may be hidden by the disk
and no eclipse can be observed. In effect the satellite is occulted by the planet
before it goes into eclipse. In general, eclipses may be observed on the western
side of Jupiter before opposition and on the eastern side after opposition. Before
opposition, only the disappearance of Satellite I into the shadow may be observed,
since it is occulted before reappearance. After opposition, only the reappearances
from the shadow are visible. In general, the same is true for Satellite II, although
occasionally both phenomena can be seen. For Satellites III and IV, both phases
36 EXPLANATORY SUPPLEMENT
of the eclipses are usually visible except near certain oppositions. Similarly, the
occultation disappearances and reappearances of the satellite cannot be observed
if at the times concerned the satellite is eclipsed.
For Satellites I and II there are usually cycles of six phenomena consisting of
both phases of both transit and shadow transit, of one phase of the eclipse, and the
other phase of the occultation. For Satellite IV, none of the phenomena occur when
the plane of the satellite's orbit, which is essentially the same as that of Jupiter's
equator, is inclined more than about 2° to the line from Jupiter to the Earth (for
occultations and transits) or to the Sun (for eclipses and shadow transits). Because
of the finite disks of the satellites, the phenomena do not take place instantaneously.
The times given refer to the centers of the disks. At certain times similar phenomena
can take place for Saturn satellites. Predictions of these phenomena are made in
special publications.
1 .56 Physical Observations of the Sun, Moon, and Planets
In addition to needing information about the positions of the bodies and the phe-
nomena due to the relative positions of the bodies, one also needs information about
the physical appearance of the bodies. A physical ephemeris gives information about
the orientation, illumination, and appearance of the body; it also provides data that
can be used in determining the cartographic coordinates of points on the surface.
For each body, it is necessary to adopt a set of rotational elements and other pa-
rameters that specify the shape of the surface and identify the position of the prime
meridian. For the Sun, and for planets and satellites whose solid surfaces cannot
be seen, it is necessary to adopt conventional values of the rotational elements. It
also may be convenient to adopt different values of the rate of rotation for certain
bands of latitude. The position of the prime meridian is fixed by adopting a value
for the longitude of the central meridian (which passes through the pole of rotation
and the apparent center of the disk) at some suitable arbitrary epoch. For a body
with a solid surface, the prime meridian is usually defined with respect to a clear,
sharp, surface feature. The appropriate ephemeris data are tabulated for the Sun,
Moon, and planets in The Astronomical Almanac. The basic values for the Sun,
Moon, planets, and satellites are given in their respective sections.
The fraction of the area of the apparent disk of the Moon or planet that is illu-
minated by the Sun is called the phase. It depends on the planetocentric elongation
of the Earth from the Sun, which is called the phase angle. Neglecting the oblate-
ness of the body, the apparent disk is circular, and the terminator is the orthogonal
projection, onto a plane perpendicular to the line of sight, of the great circle that
bounds the illuminated hemisphere of the body. The terminator is therefore, in
general, an ellipse, reducing to a straight line at a phase angle of 90° and becoming
a circle at a phase angle of 0° or 1 80° .
1 / INTRODUCTION TO POSITIONAL ASTRONOMY
37
The elements of the rotational motion are the period of rotation, the position of
the axis of rotation in space (represented either by the coordinates of the point on
the celestial sphere toward which the axis is directed; i.e., the pole of the rotation,
or by the inclination and node of the equator on an adopted reference plane) , and
the planetographic longitude of the central point of the apparent disk at an adopted
epoch. This longitude defines the central meridian on the disk.
In general, the apparent positions of points on the disk are represented most
conveniently by the apparent distance and position angle relative to the central
point of the disk. Position angles are ordinarily measured from the north point of
the disk toward the east. The central point of the apparent disk is the subterrestrial
point on the surface. Its position on the geocentric celestial sphere is diametrically
opposite the apparent position of the Earth on the planetocentric celestial sphere.
The north point of the disk is on the apparent northern limb at its intersection with
the celestial meridian that, passes through the north celestial pole and the center
of the disk. The vertex of the disk is on the apparent upper limb at its intersection
with the vertical circle that passes through the zenith and the center of the disk.
Two types of coordinate systems are used for physical observations. Planeto-
centric coordinates are for general use where the s-axis is the mean axis of rotation,
the jc-axis is the intersection of the planetary equator (normal to the z-axis through
the center of mass) and an arbitrary prime meridian, and the y-axis completes a
right-hand coordinate system. Longitude of a point is measured positive to the east
from the ephemeris position of the prime meridian as defined by rotational ele-
ments. Latitude of a point is the angle between the planetary equator and a line to
the center of mass. The radial distance is measured from the center of mass to the
surface point.
Planetographic coordinates are used for cartographic purposes and are depen-
dent on an equipotential surface or an adopted ellipsoid as a reference surface.
Longitude of a point is measured in the direction opposite to the rotation (pos-
itive to the west for direct rotation) from the cartographic position of the prime
meridian defined by a clearly observable surface feature. Latitude of a point is the
angle between the planetary equator (normal to the z-axis and through the center
of mass) and the normal to the reference surface at the point. The height of a point
is specified as the distance above a point with the same longitude and latitude on
the reference surface.
38
EXPLANATORY SUPPLEMENT
1.6 REFERENCES
Brouwer, D. and Clemence, G.M. (1961). Methods of Celestial Mechanics (Academic Press,
New York and London).
Danby, J.M.A. (1962, 1988). Fundamentals of Celestial Mechanics (Willmann-Bell, Inc.,
Richmond, VA).
Explanatory Supplement to The Astronomical Ephemeris and The American Ephemeris
and Nautical Almanac (1961). (Her Majesty's Stationery Office, London).
Hagahara, Y. (1970). Celestial Mechanics Volumes I and II (MIT Press, Cambridge, MA
and London, England) Volumes III, IV and V (Japan Society for the Promotion of
Science, Kojimachi, Cheyoda-ku, Tokyo).
Herget, P. (1948). Computation of Orbits Published privately by the author. (Willmann-
Beli, Inc., Richmond, VA).
Mueller, LI. (1969). Spherical and Practical Astronomy as Applied to Geodesy (Ungar,
New York).
Woolard, E.W. and Clemence, G.M. (1966). Spherical Astronomy (Academic Press, New
York and London).
CHAPTER
Time
by P.K. Seidelmann, B. Guinot, and LE. Doggett
2.1 INTRODUCTION
The methods and accuracy of timekeeping have changed dramatically since 1960.
The definition of the second — the most precisely reproducible unit in the Systeme
International (SI) — was once based on the rotation or the motion of the Earth. Now
the second is based on natural frequencies of atoms of selected elements. These
changes are significant for three reasons:
(1) All time systems now refer to the SI second.
(2) Precise time is available on a real-time basis.
(3) The unit of distance, the meter, has been defined most precisely in terms
of the distance traveled by light in a ^iven time interval.
To establish a system of time, one must define two quantities: the unit of
duration (for example, the second or the day), and the epoch, or the zero, of the
chosen time. In physics and astronomy, there are four types of systems in common
use. Broadly speaking, they are the following:
(a) Atomic time, in which the unit of duration corresponds to a defined number
of wavelengths of radiation of a specified atomic transition of a chosen
isotope.
(b) Universal Time, in which the unit of duration represents the solar day,
defined to be as uniform as possible, despite variations in the rotation of
the Earth.
(c) Sidereal time, in which the unit of duration is the period of the Earth's
rotation with respect to a point nearly fixed with respect to the stars.
(d) Dynamical time, in which the unit of duration is based on the orbital motion
of the Earth, Moon, and planets.
39
40 EXPLANATORY SUPPLEMENT
Fast-rotating pulsars (millisecond pulsars) may in the future lead to a new time
system. In each of these systems, various definitions of the unit of duration and of
the epoch have prevailed in different periods of recent astronomical history. The
following section describes the details of these systems.
2.2 MEASURES OF TIME AND THEIR RELATIONSHIPS
2.21 Atomic Time (TAI)
Atomic time is based on counting the cycles of a high-frequency electrical sig-
nal that is kept in resonance with an atomic transition. The fundamental unit of
atomic time is the Systeme International (SI) second. It is denned as the duration
of 9,192,631,770 periods of the radiation corresponding to the transition between
two hyperfine levels of the ground state of the cesium-133 atom (see Section 2.312).
International Atomic Time (Temps Atomique International, or TAI) is a prac-
tical time standard that conforms as closely as possible to the definition of the
SI second. It fulfills the needs of accuracy, long-term stability, and reliability for
astronomy. Although TAI was officially introduced in January 1972, it has been
available since July 1955. The SI second and International Atomic Time are used
as bases for interpolation and prediction of the other timescales.
TAI results from the analysis by the Bureau International des Poids et Measures
in Sevres, France, of individual timescales from commercial atomic-time standards
and the primary frequency standards in many countries. It is realized in two steps.
First, an intermediate time scale, denoted EAL (Echelle Atomique Libre), is formed
by combining data from all available high-precision atomic standards. The data
from the individual clocks are combined using an algorithm that minimizes the
effect of adding and dropping clocks and that weights clocks appropriately. As the
name implies, EAL is a free-running, data-controlled timescale. After analysis of
EAL, corrections are applied for known effects to maintain a unit of time as close as
possible to the definition of the SI second. This adjusted timescale is published as
TAI. Since TAI is accessible through the published corrections to each contributing
timescale, it is often called a paper-scale clock.
A large number of commercial cesium standards and a few hydrogen masers
provide availability, reliability, and short-term stability to the TAI system. Accuracy
and long-term stability are provided by laboratory cesium clocks or standards.
It is anticipated that improvements in the clocks or standards will permit future
improvements in the International Atomic Time, but the fundamental principles
underlying the formulation of the TAI will remain the same.
2.211 Relativistic Effects Advances in the accuracy of time comparisons require
the adoption of a set of conventions and a coordinate reference frame to account
for relativistic effects in an internally consistent manner. At its 9th session (1980
2 / time 41
September 23-25), the Consultative Committee for the Definition of the Second
(CCDS) recognized this need and proposed the following to the International Com-
mittee of Weights and Measures (CIPM):
(a) that TAI is a coordinate timescale defined at a geocentric datum line and
having as its unit one SI second as obtained on the geoid in rotation;
(b) that with the present state of the art, TAI may be extended with sufficient
accuracy to any fixed or mobile point near the geoid by applying the correc-
tions of the first-order of general relativity (i.e., corrections for differences
in gravitational potential and velocity and for the rotation of the Earth).
2.22 Dynamical Time
Dynamical time represents the independent variable of the equations of motion of
the bodies in the solar system. According to the theory of relativity, this inde-
pendent variable depends on the coordinate system being used as the system of
reference. In modern astronomy, the equations of motion are often referred to the
barycenter of the solar system. As a result there is a family of barycentric dynami-
cal timescales that depend on various forms of relativistic theory. The independent
variable of an apparent geocentric ephemeris is a terrestrial dynamical time, which
is also theory dependent. For a given relativistic theory there exists a transforma-
tion between the barycentric and terrestrial dynamical timescales. The arbitrary
constants in the transformation can be chosen so that the timescales have only
periodic variations with respect to each other.
Since the transformation between barycentric and geocentric timescales de-
pends upon the theory, the two types of scales cannot both be unique. Thus,
the dynamical timescale for apparent geocentric ephemerides was chosen to be
a unique timescale, independent of the theories, whereas the dynamical timescales
for barycentric ephemerides form a family of timescales that are not unique and are
theory-dependent.
The recommendations for the definition of dynamical time adopted at the IAU
General Assembly of 1976 in Grenoble, France are as follows:
(a) At the instant 1977 January OIWOXTOO* TAI, the value of the new timescale
for apparent geocentric ephemerides will be 1977 January 1?0003725
(l"00h0(r32!184) exactly.
(b) The unit of this timescale will be a day of 86,400 SI seconds at mean sea
level.
(c) The timescales for equations of motion referred to the barycenter of the solar
system will be such that there will be only periodic variations between these
timescales and those of the apparent geocentric ephemerides.
42 EXPLANATORY SUPPLEMENT
In 1991, in the framework of resolutions encompassing both space and time
references, the IAU has defined new timescales TCB, TCG, and TT, as explained
in Section 2.223. In the 1976 and 1991 resolutions, the dynamical time of planetary-
motions is identical to the timescale of terrestrial atomic physics. The consequences
of this assumption are considered in Section 2.34.
2.221 Terrestrial Dynamical Time Terrestrial Dynamical Time (TDT) is the the-
oretical timescale of apparent geocentric ephemerides of bodies in the solar system.
TDT is specified with respect to TAI in order to take advantage of the direct avail-
ability of Coordinated Universal Time (UTC), which is based on the SI second.
TDT is currently set equal to TAI + 32,184 seconds.
Since TAI is a statistical timescale based on a large number of clocks operat-
ing on the Earth, and TDT is an idealized uniform timescale, the two timescales
cannot be made identical. TAI is subject to systematic errors in the length of the
TAI second and in the method of forming TAI. For the near future, however, the
cumulative effect of such errors is likely to be significant only for evaluation of the
millisecond pulsars.
The relationship between TDT and TAI provides continuity with ephemeris
time (ET). The chosen offset between TDT and TAI is equal to the estimate of the
difference between ET and TAI at the time that TDT was introduced. For years
prior to 1955, when atomic time was not available, TDT must be extrapolated
backward, using the theory of a rapidly moving celestial body that has been fit to
observations made since 1955. The Moon and Mercury are the most suitable bodies
for such purposes. The accuracy of the backward extrapolation will depend on the
length of the comparison after 1955 and will deteriorate as a function of the length
of time prior to 1955. Use of the Moon for this extrapolation is complicated by the
Moon's tidal acceleration. At this time there are uncertainties concerning the value
of the tidal acceleration and the systematic variations between the theoretically
calculated position of the Moon and the observations.
2.222 Barycentric Dynamical Time Barycentric Dynamical Time {TDB) is the in-
dependent variable of the equations of motion with respect to the barycenter of the
solar system. In practice, TDB is determined from TDT by means of a mathe-
matical expression. This expression depends upon the constants, the positions and
motions of the bodies of the solar system, and the gravitational theory being used.
An approximate formula is sufficient in most cases for converting from TDT to
TDB. Such a formula is
TDB = TDT + 0!00 1 658 sin g + 0!0000 1 4 sin 2g , (2.222- 1 )
where
g = 357?53 + 0, 9856003(JD - 2451545.0)
2 / TIME 43
and JD is the Julian date to two decimals of a day.
A more accurate expression is given by Moyer (1981):
TDB - TDT = AT a + 4 VW0<? sin E + , sinD
c1 c2(l + u)
Sr{\ + cose)«
+ 2c2 X
(1 - ^e2) sin(UTl + A) + e sin(UTl +X-M)
9 ,
+ -e2 sin(UTl +A - 2M)
- Scil ~^°S£)M[sin(UTl + A + 2L) + esin(UTl + A + 2L + M)]
2cl
Sm{\ +cose)u .
+ o ^/u x sin(UTl +X-D)
2c2(l +n)
Sc(sine)v iijej , Msa^sa . _
s cosL + -~ sin£j + — sin£SA
+ Sm(-L ~ Lti + — Sln^ ~ LSA)
c2fi* czn*
+ MJ&Cl+cosg) T1 + A+L_4)
2cl /i*
^sa5sa(1 +cose)w . L\J_rr\ mil t,
+ — - - sin(UTl + A + L - LSa). (2.222-2)
2c 2fi*
When numerical values are substituted into the expression, the relationship
becomes:
TDB — TDT - A7X + 1 . 658 x 1CT3 sin£+ 1.548 x 10"6sin£»
+ 3. 17679 x 10_1(Wn(UTl +A) + 5.312x lCTI2u sin(UTl + A - M)
+ 1.00 x 10_13wsin(UTl + A-2M) - 1.3677 x 10~UHsin(UTl + A + 2L)
- 2. 29 x 1 0~uu sin(UTl + A + 2 L + A/) + 1 . 33 x 10~ l3u sin(UTl + A - D)
- 1.3184 x 10-'°vcosL + 5.21 x 10-6sinEj
+ 2. 45 x 10"6 sin £SA + 20. 73 x 10"6 sin(L - Ls)
+ 4. 58 x 10"6 sin(L - LSk) + 1. 33 X 10" !3w sin(UTl + A + L - L3)
+ 2.9 x 10_14Hsin(UTl + A + L - LSA). (2.222-3)
The symbols in Equations 2.222-2 and 2.222-3 are defined as follows:
A7"A — constant term in expression
c = speed of light
r|, rf = position and velocity vectors of point i relative to point j. The dot denotes
differentiation with respect to coordinate time t
44 EXPLANATORY SUPPLEMENT
Hi = gravitational constant of body i
fi* - gravitational constant of the Sun augmented by the gravitational con-
stants of the planets and the Moon
a,-, ei, Ej = semi-major axis, eccentricity, and eccentric anomaly, respectively, of
heliocentric orbit of planet i; without subscript, the orbit of Earth-Moon
barycenter
St = circular orbit velocity of planet i relative to Sun, i/(fis + A*;) / ai
A = east longitude of atomic clock
u = distance from Earth's spin axis of atomic clock in km
v = coordinate of atomic clock in distance from Earth's equatorial plane
(positive north) in km
am = semi-major axis of geocentric orbit of the Moon
Sm = circular orbit velocity of the Moon relative to Earth, V(Me + fj,m) I am
e = mean obliquity of the ecliptic, i.e., inclination of the ecliptic plane to
mean Earth equator of date
<?m = Greenwich mean sidereal time = Greenwich hour angle of mean equinox
of date
5c - circular orbit velocity of Earth-Moon barycenter relative to the Sun,
V(j"s + Me + Mm) / a
L = mean longitude of the Sun with respect to the Earth-Moon barycentric
referred to the mean equinox and ecliptic date
D - the mean elongation of the Moon from the Sun
Li — heliocentric mean longitude of planet /
Subscripts and superscripts are as follows:
A - location of atomic clock on Earth that reads International Atomic Time (r)
rn - Moon
S = Sun
C = solar system barycenter
E = Earth
J = Jupiter
SA = Saturn
The absence of a subscript indicates the Earth.
The relation between the eccentric and mean anomalies is:
E = M + e sin M + (e2 I 2) sin 2 M + - • . (2.222-4)
The eccentric anomalies of the Earth, Jupiter and Saturn can be computed to
sufficient accuracy from
£e = Me + e sin A/e
2 / TIME
45
Es = Mj (2.222-5)
Esk = MSa
2.223 Space-Time Coordinates In 1991 the IAU adopted resolutions clarifying the
relationships between space-time coordinates. These resolutions introduced Terres-
trial Time (TT), Geocentric Coordinate Time (TCG), and Barycentric Coordinate
Time (TCB), and they permitted the continued use of Barycentric Dynamical Time
(TDB).
The space-time coordinates (x° = ct, x1, X2, X3) would be defined in the frame-
work of the general theory of relativity in such a way that in each coordinate system
centered at the barycenter of an ensemble of masses exerting the main action, the
interval ds2 would be expressed at the minimum degree of approximation by
ds2 = -<*dTi = -fl-^j (dx°f+(l + ^) [(^x])2 + (^2)2 + (^3)2] (2.223-1)
where c is the velocity of light, r is the proper time, and U is the sum of the
gravitational potentials of the ensemble of masses and a tidal potential generated
by the external bodies, the latter potential vanishing at the barycenter.
The space-time coordinates are further constrained such that:
(1) The space coordinate grids with origins at the solar system barycenter and
at the Earth's center of mass show no global rotation with respect to a set
of distant extragalactic objects.
(2) The time coordinate is derived from a timescale realized by atomic clocks
operating on the Earth.
(3) The basic physical units of space-time in all coordinate systems are the SI
second for proper time and the SI meter for proper length (this is related
to the SI second by the value of the velocity of light: c = 299792458ms"1).
It should be noted that the preceding kinematic constraint for the state of ro-
tation of the geocentric reference system implies that when the system is used for
dynamics (e.g., the motion of the Moon), one must take into account the time-
dependent geodesic precession in the motion of the Earth, and introduce the corre-
sponding inertial terms in the equations. Also, if the previously defined barycentric
reference system is used for dynamical studies, the kinematic effects of the galactic
geodesic precession may have to be taken into account.
To be consistent with the proposed space-time coordinates, coordinate times
would be specified such that:
(1) The units of measurement of the coordinate times of all coordinated systems
centered at the barycenters of ensembles of masses are chosen so that they
are consistent with the proper unit of time, the SI second.
46
EXPLANATORY SUPPLEMENT
(2) The reading of these coordinate times is 1977 January I, 0h0m32M84 exactly,
on January 1, 0h0m0B TAI exactly (JD = 2443144.5, TAI), at the geocenter.
(3) Coordinate times in coordinate systems that have their spatial origins at
the center of mass of the Earth and at the solar system barycenter and
that meet the previously stated requirements are designated as Geocen-
tric Coordinate Time (TCG) and Barycentric Coordinate Time (TCB),
respectively.
TCG and TCB were adopted by the IAU in 1991. They are in a way com-
plementary to TDT and TDB. but it must be recognized that some astronomical
constants and quantities have different numerical values when using TDB, as op-
posed to TDT, TCG, or TCB. It should further be noted concerning TCB and
TCG that:
(1) In the domain common to any two coordinate systems, the tensor transfor-
mation law applied to the metric tensor is valid without repealing the unit
of time. Therefore the various coordinate times exhibit secular variations.
Recommendation 5 (1976) of IAU Commissions 4, 8, and 31, completed
by Recommendation 5 (1979) of IAU Commissions 4, 19, and 31, stated
that Terrestrial Dynamical Time (TDT) and Barycentric Dynamical Time
(TDB) should differ only by periodic variations. Therefore TDB and TCB
differ in rate. The relationship between these timescales in seconds is given
by:
TCB - TDB = LB x (JD - 2443144. 5) x 86400 (2.223-2)
The present estimate of the value of LB is 1. 550505 x 10~s (+ 1 x tO"14).
(2) The relation TCB — TCG involves a full four-dimensional transformation:
TCB - TCG = c~2
f
J to
(vl/2+ E/ext(xE)) dt + vc ■ (x - xe)
(2.223-3)
Xf and ve denote the barycentric position and velocity of the Earth's center
of mass, and X is the barycentric position of the observer. The external
potential, Utu, is the Newtonian potential of all solar system-bodies apart
from the Earth. The external potential must be evaluated at the geocenter.
In the integral, / = TCB and is chosen to agree with the epoch of Ter-
restrial Time (TT). As an approximation to TCB — TCG in seconds one
might use:
TCB -TCG = Lc x (JD- 2443 144. 5) x 86400+c^2ve-(x-xe)+/". (2.223-4)
The present estimate of the value of Lc is 1.480813 x 10"8 (± 1 x 10-14).
The quantity P represents the periodic terms which can be evaluated using
analytical formulas. For observers on the surface of the Earth, the terms
2 / TIME
47
depending upon their terrestrial coordinates are diurnal, with a maximum
amplitude of 2.1 (is.
(3) The origins of coordinate times have been arbitrarily set so that these times
all coincide with the Terrestrial Time (TT) at the geocenter of 1977 January
1, 0"0m0a TAI.
(4) When realizations of TCB and TCG are needed, it is suggested that these
realizations be designated by expressions such as TCB(jccr), where xxx in-
dicates the source of the realized timescale (e.g., TAI) and the theory used
for transformation into TCB or TCG.
In addition, it was agreed to delete the word "dynamical" from Terrestrial
Dynamical Time (TDT) and just call it Terrestrial Time (TT). TT would be defined
as follows:
(1) The time reference for apparent geocentric ephemerides is Terrestrial Time,
TT.
(2) Terestrial Time is a timescale differing from TCG uniquely by a constant
rate, its unit of measurement being chosen so that it agrees with the SI
second on the geoid.
(3) At instant 1977 January 1, 0h0m0s TAI exactly, TT has the reading 1977
January 1, 0h0m32!184 exactly.
It should be noted concerning TT that:
(1) The divergence between TAI and TT is a consequence of the physical de-
fects of atomic time standards. In the interval 1977-1990, in addition to
the constant offset of 32! 184, the deviation probably remained within the
approximate limits of ± 10 (is. It is expected to increase more slowly in the
future as a consequence of improvements in time standards. In many cases,
especially for the publication of ephemerides, this deviation is negligible.
In such cases, it can be stated that the argument of the ephemerides is
TAI + 32! 184.
(2) Terrestrial Time differs from TCG in seconds uniquely by a scaling factor:
TCG - TT = Lc x ( JD - 2443 144. 5) x 86400 (2.223-5)
The present estimate of the value of LG is 6.969291 x 10"10 (± 3 x 10"16).
The two timescales are distinguished by different names to avoid scaling
errors.
(3) The unit of measurement of TT is the SI second on the geoid. The usual
multiples, such as the TT day of 86400 SI seconds on the geoid, and the TT
Julian century of 36525 TT days can be used, provided that the reference
to TT be clearly indicated whenever ambiguity may arise. Corresponding
time intervals of TAI are in agreement with the TT intervals within the
48 EXPLANATORY SUPPLEMENT
uncertainties of the primary atomic standards (e.g., within + 2 x 10^ 14 in
relative value during 1990).
(4) It is suggested that realizations of TT be designated by TT(xtr), where xxx
is an identifier. In most cases, a convenient approximation is:
TT(TAI) = TAI + 32? 184 (2.223-6)
However, in some applications it may be advantageous to use other realiza-
tions. The BIPM, for example, has issued timescales such as TT (BIPM90)
The concepts were adopted by IAU resolution and will probably be implemented
into general use in the future. There is an Astronomical Standards Working Group
currently considering various details of interpretation and implementation.
2.23 Sidereal Time
In general terms, sidereal time is the hour angle of the vernal equinox, which is also
known as the first point of Aries. Apart from the inherent motion of the equinox, due
to precession and nutation, sidereal time is a direct measure of the diurnal rotation
of the Earth. Equal intervals of angular motion correspond to equal intervals of
mean sidereal time, so that sidereal time reflects the actual rotation of the Earth,
and can be determined by observations of celestial objects.
The sidereal time measured by the horn angle of the true equinox — i.e., the
intersection of the true equator of date with the ecliptic of date — is apparent sidereal
time. The position of the true equinox is affected by the nutation of the axis of
the Earth, which consequently introduces periodic inequalities into the apparent
sidereal time. The time measured by the diurnal motion of the mean equinox of
date, which is affected by only the secular inequalities due to the precession of
the axis, is mean sidereal time. Apparent sidereal time minus mean sidereal time
is the equation of the equinoxes. In ephemerides immediately preceding 1960, the
equation of the equinoxes was called the nutation in right ascension.
The period between two consecutive upper meridian transits of the equinox is a
sidereal day. It is reckoned from 0h at upper transit, which is known as sidereal noon.
Owing to precession, the mean sidereal day of 24 hours of mean sidereal time is
shorter than the actual period of rotation of the Earth by about 0!0084, the amount
of precession in right ascension in one day. The ratio of the length of the mean
sidereal day to the period of rotation of the Earth is 0. 999999902907 — 59 x 10-12 T,
where T is measured in Julian centuries from J2000.0. Correspondingly, the period
of rotation is 1.000000097093 + 59 Tx 10 12 mean sidereal days. These numbers are
not rigorously constant because the sidereal motion of the equinox due to precession
is proportional to the length of the UT day, that is, to the period of the rotation
of the Earth, whereas the angular measure of the complete rotation is, of course,
2 / TIME
49
constant. However, the conceivable change in the period of rotation is such that the
effect of a variation in the daily precessional motion is inappreciable. The secular
variations are almost inappreciable (see Section 3.21). The apparent sidereal day,
nominally of 24 hours of apparent sidereal time, differs from the period of rotation
by a variable amount depending on the nutation.
To each local meridian on the Earth there corresponds a local sidereal time,
connected with the sidereal time of the Greenwich meridian by means of the relation:
local sidereal time = Greenwich sidereal time + east longitude. (2.23-1)
Sidereal time is conventionally measured in hours, minutes, and seconds, so that
longitude is measured (positively to the east) in time at the rate of one hour to 15°.
An object transits over the local meridian when the local sidereal time is equal to
its right ascension.
Classically, sidereal time was determined from observations of the transits of
stars, either over the local meridian or, with a prismatic astrolabe, over the small
circle corresponding to a constant altitude. Because the reduction of these observa-
tions depended on star positions taken from a standard star catalog, sidereal time
was actually based on the hour angle of the zero point of right ascension of the
catalog, known as the catalog equinox. This point is distinct from (but close to) the
dynamical equinox defined by the orbital motion of the Earth (see Section 3.1).
Apparent sidereal time, because of its variable rate, is used only as a measure
of epoch; it is not used as a measure of time interval. Observations of the diurnal
motions of the stars provide a direct measure of apparent sidereal time, as their
right ascensions are measured from the true equinox. But in many practical methods
of determining time, the right ascensions are diminished by the equation of the
equinoxes, so that mean sidereal time is deduced directly from the observations.
Today, sidereal time is determined most accurately by interferometer obser-
vations of radio sources. It can also be determined from lunar and satellite laser-
ranging. Since the rotation of the Earth is subject to irregular forces, sidereal time
is irregular with respect to atomic time. The practical determination of sidereal
time requires that short-period irregularities and polar motion be evaluated at the
same time.
2.231 Greenwich Sidereal Date In order to facilitate the enumeration of succes-
sive sidereal days, the concepts of Greenwich sidereal date and Greenwich sidereal
day number, analogous to those of Julian date and Julian day number (see Sec-
tion 2.26), have been introduced. The Greenwich sidereal date is defined as the
interval in sidereal days, determined by the equinox of date, that has elapsed on
the Greenwich meridian since the beginning of the sidereal day that was in progress
at JD 0.0. The integral part of the Greenwich sidereal date is the Greenwich sidereal
day number, a count of transits of the equinox over the Greenwich meridian. The
50 EXPLANATORY SUPPLEMENT
decimal part of the Greenwich sidereal date is simply the Greenwich sidereal time
expressed as a fraction of a sidereal day. These concepts can be applied equally well
to mean or apparent sidereal time.
2.24 Universal Time
Universal Time (UT) is the measure of time used as the basis for all civil time-
keeping. It conforms closely to the mean diurnal motion of the Sun. The apparent
diurnal motion of the Sun involves both the nonuniform diurnal rotation of the
Earth and the motion of the Earth in its orbit around the Sim. Although it would
be possible to define a system of time measurement in terms of the hour angle of
the Sun, such a system could never be related precisely to sidereal time and could
not, therefore, be determined by observations of star transits. As a result, Universal
Time is directly related to sidereal time by means of a numerical formula. It does
not refer to the motion of the Earth and is not precisely related to the hour angle
of the Sun.
Universal Time at any instant can be derived from observations of the diurnal
motion of the stars or radio sources. The uncorrected observed rotational timescale,
which is dependent on the place of observation, is designated UTO. Correcting this
t imescale for the shift in longitude of the observing station caused by polar motion
produces the UTl timescale, which is independent of observing location, but is
influenced by the slightly variable rotation of the Earth.
Since 1984 January 1 Greenwich mean sidereal time (GMST) has been related
to UTl as follows:
GMSTI of 0hUTJ =24110!54841 +86401 84?8 12866 Ta+ fr!093104 7^ -6.2 x 1(T67^
(2.24-1)
where Tu - du I 36525, du is the number of days of Universal Time elapsed since
JD 2451545.0 UTl (2000 January 1, 12hUTl), taking on values of ± 0.5, ± 1.5,
± 2. 5, ± 3. 5, This relation conforms to the position and motion of the equinox
specified in the IAU 1976 System of Astronomical Constants, the 1980 IAU Theory
of Nutation, and the positions and proper motions of stars in the FK5 catalog.
Equation 2.24-1 is often considered to be the definition of UTl, because when
it was established, the observation of transits of the FK5 stars (or of other stars
with their positions and proper motions in the FK5 system) was the best means
to obtain UTl. When new techniques for measuring UTl became operational (see
Section 2.32), the arbitrary constants involved in their implementation were set to
align I he 111 values with those obtained by observations of FK5 stars and to be
consistent with Equation 2.24 1.
The modern definition of UTl was established in order to fulfill the following
conditions:
2 / TIME 51
(a) UTl is proportional to the angle of rotation of the Earth in space, reck-
oned around the true position of the rotation axis. In other words, UTl is
proportional to the integral of the modulus of the rotation vector of the
Earth.
(b) The rate of UTl is chosen so that the day of UTl is close to the mean
duration of the solar day.
(c) The phase of UTl is chosen so that 12h UTl corresponds approximately, in
the average, to the instant when the Sun crosses the meridian of Greenwich.
Conditions (b) and (c) are not strictly compatible with condition (a). By adopt-
ing (a), one accepts that UTl deviates secularly from solar time; however, the
divergence is extremely small.
The constant term and the numerical coefficient of 7"„ in Equation 2.24-1 were
chosen so that UTl would be continuous in value and rate through the change from
the old to the new system of constants (1984 January 1). The small differences due
to the change in the theory of nutation were neglected.
The variation of UTl with respect to uniform time truly represents the irreg-
ularities of the rotation of the Earth. The duration (A) of the day of UTl in SI
seconds is given by
A = 86400 n, (2.24-2)
where Vi and are the values of UTl - TAI in seconds at n-day intervals. The
angular velocity of the Earth, w, is given by
u = (86400 / A)x 72. 921 15 1467 x 10_6rads_1. (2.24-3)
For the dynamics of the solar system, the study of the rotation of the Earth, and the
study of the Galaxy, one needs the relationship between the position and motion
of the equinox and those of a point on the equator, the hour angle of which truly
represents the rotation of the Earth in space (sidereal rotation). There are several
ways to establish this relationship, including using statistics on the proper motions
of stars, the dynamical method, and observations of quasars; these lead to slightly
different realizations of the equinox and, therefore, of the sidereal time. Therefore,
an agreement has been made to take as a reference the equinox and its motion as
realized by the IAU 1976 System of Astronomical Constants, the 1980 IAU Theory
of Nutation, and the positions and proper motions of stars of the FK5 catalog.
2.241 Conversions between Universal Time and Sidereal Time The number of sec-
onds of sidereal time in a Julian century can be obtained by direct differentiation
of Equation 2.24-1:
s' = 8640184.812866 + 0. 1862087- ].86x 10~5 T2.
(2.241-1)
52 EXPLANATORY SUPPLEMENT
Dividing by 36525 (the number of days in a Julian century) gives the number of
seconds of mean sidereal time in a day of Universal Time (UT1 to be precise):
s = 86636*55536790872 + 5. 098097 x 10~6r - 5. 09 x 10" ,0T2 (2.241 2)
where T is the number of Julian centuries elapsed since JD 2451545.0. The quantity
s represents the number of sidereal seconds in a day of Universal Time. Dividing
Equation 2.241-2 by 86400 gives the ratio of mean sidereal time to UTl:
r' = 1.002737909350795 + 5.9006 x 10_11r- 5.9 x 10_157;2. (2.241-3)
Taking the inverse of this expression gives the ratio of UTl to mean sidereal time:
I // = 0.997269566329084 -5. 8684 x 10 1 ]T+5. 9 x 10" l5T2. (2.241-1)
Although the lengths of a day of UTl and a day of mean sidereal time vary
slightly with variations in the Earth's rotation, the ratio of UTl to mean sidereal
time is given by Equation 2.241-4 and is unaffected by rotational variations. Thus,
the established method of determining Universal Time by multiplying an interval
of sidereal time elapsed from 0h UT by a fixed conversion factor maintains the
constancy of the ratio of Universal Time to sidereal time, irrespective of variations
in the rate of rotation of the Earth.
Roughly speaking, UTl can be regarded as solar time or dynamical time in
these equations. Then Equation 2.241-4 can be thought of as giving the length
of one sidereal day in units of solar days. Disregarding the inappreciable secular
variations, the equivalent measures of the lengths of the days are:
mean sidereal day = 23h56m04?090524 of Universal Time,
day of Universal Time - 24"03m56*5553678 of mean sidereal time.
The rotational period of the Earth is 86164.09890369732 seconds of UTl or
23h56mO4!09890369732. This gives a ratio of a UTl-day to the period of rotation
of 1.00273781 1906. The rate of rotation is 15.04106717866910 seconds of arc per
second of time.
2.242 Conversion of Local Apparent Sidereal Time to UTl The usual procedure
to express the UTl date of an event that occurred at some instant dated in local
apparent sidereal time (LAST) is as follows:
(a) Express the date in local mean sidereal time (LMST) by
LMST = LAST - EQ,
(2.242-1)
2 / TIME
53
where EQ is the equation of the equinoxes.
(b) Express the date in Greenwich mean sidereal time by
GMST1 = LMST - A, (2.242-2)
where A is the astronomical longitude of the instrument, positive toward
the east. Since the pole of rotation is moving with respect to the surface
of the Earth, the instrumental meridian is subject to small variations. The
position of the rotational pole (jc,y) with respect to the Conventional Inter-
national Origin (CIO) can be used to correct for this effect. This correction
of Ao (referred to CIO) to A is accomplished by applying the expression
A = A0 - (x sin A0 + y cos A0) tan <p (2.242-3)
where 4> is the latitude of the instrument.
(c) Compute the time 9 elapsed between 0h UTl and the event, expressed in
mean sidereal time by
9 = GMST1 - GMST1 of 0hUTl . (2.242-4)
(d) Convert 9, expressed as an interval of sidereal time, to an interval of UTl:
UTl = rd (2.242-5)
where r is the ratio of mean solar time to sidereal time intervals, given in
Equation 2.241-4.
For instance when the hour angle, H, of a star with right ascension a has been
measured, the corresponding UTl value is
UTl = (H + a-EQ-X- GMST1 of 0hUTI)r. (2.242-6)
The relationship between LMST and local civil time (LCT) is given by
LMST of 0" LCT = GMST1 of 0" UTl - (1 - r)A / r, (2.242-7)
where LCT = UTl + A, and GMST1 of 0" UTl is obtained from Equation 2.24-1.
2,243 Coordinated Universal Time Since 1972 January 1, most, and now all, broad-
cast time services distribute timescales based on the redefined Coordinated Univer-
sal Time (UTC), which differs from TAI by an integer number of seconds. UTC is
maintained within 0.90 second of UTl by the introduction of one-second steps (leap
54 EXPLANATORY SUPPLEMENT
seconds) when necessary, usually at the end of June or December. DUT1, an ex-
trapolation of the difference UTl minus UTC, is transmitted in code on broadcast
time signals.
2.25 The Ephemeris Meridian
Since hour angles are calculated with reference to the meridian of Greenwich or
other geographic meridians, they are dependent on the rotation of the Earth. For
solar-system bodies, whose motions are expressed as a function of dynamical time,
such calculations can only be carried out accurately after a value of AT" ( = TDT —
UT) is known. To facilitate practical calculations of phenomena that depend upon
hour angle and geographic location, the concept of an auxiliary reference meridian,
known as the ephemeris meridian, was introduced.
Although the ephemeris meridian was originally used in conjunction with ephem-
eris time, its name has been retained for use with calculations in TDT. Its position
in space is conceived as being where the Greenwich meridian would be if the Earth
rotated uniformly at the rate implicit in the definition of dynamical time; i.e.,
1.002738A77 east of the actual meridian of Greenwich on the surface of the Earth,
Calculations in terms of TDT may be referred to the ephemeris meridian us-
ing methods that are formally the same as those done in UT, when referred to
the Greenwich meridian. As soon as a sufficiently accurate value of AT can be
determined or extrapolated, the longitudes and hour angles can be referred to the
Greenwich meridian, and the times in TDT can be expressed in UT. This procedure
is followed, for example, in predictions of the general circumstances of eclipses.
Equation 2.24-1 gives the hour angle of the equinox at 0h UT, referred to the
actual geographic meridian of Greenwich. However, if T is expressed in dynamical
time rather than Universal Time, Equation 2.24-1 gives the hour angle of the
equinox at 0h TDT, referred to the ephemeris meridian.
The Universal Time of transit of the Sun, Moon, or a planet across the meridian
of Greenwich can be found by subtracting AT from the dynamical time of Green-
wich transit. To determine the dynamical time of Greenwich transit, the tabulated
time of transit across the ephemeris meridian must be interpolated from the ge-
ographic longitude of the ephemeris meridian (1.002738 AT east) to longitude 0°.
The Universal Time of Greenwich transit is given approximately by
UT of Greenwich transit = TDT of ephemeris transit + (AT / h)Aa, (2.25-1)
where Aa is the first difference in tabular right ascension and h is the tabular
interval.
It is possible to define hour angles, sidereal time, and an equation of time with
respect to the ephemeris meridian. But to relate such a system to observations
from the surface of the Earth, the value of AT" is required. It is more convenient
2 / TIME
55
Ephemeris meridian Local meridian True Sun Equinox
1.002738 AT
k — )
12h +UT
— >
Equation of time
3cal hour angle of Sun
>
Right ascension of Sun
Longitude
Hour angle of Sun
L
s
Sidereal time
!
Universal meridian Mean Sun of UT
(Greenwich)
Figure 2.25.1
Meridian relations and time
and conventional to convert TDT to UT by means of AT and to relate hour angles,
longitudes, and times to the Greenwich meridian.
The speed of rotation of the ephemeris meridian is such that it makes one
complete revolution of 360°, relative to the mean equinox, in 23h56m04!098904 of
dynamical time. The ephemeris meridian coincided with the Greenwich meridian
at some date between 1900 and 1905.
Apart from its practical advantages, the concept of the ephemeris meridian
is valuable in providing a clear picture of the relation between dynamical time
and Universal Time. Figure 2.25.1 shows the relationship between the ephemeris
meridian and the universal (or Greenwich) meridian. It also shows the relationships
between the local meridian, the equinox, and the true and mean Sun.
2.26 Julian Date
To facilitate chronological reckoning, astronomical days, beginning at Greenwich
noon, are numbered consecutively from an epoch intended to be sufficiently far in
the past to precede the historical period. The number assigned to a day in this
continuous count is the Julian day number, which is defined to be 0 for the day
starting at Greenwich mean noon on 1 January 4713 B.C., Julian proleptic calendar.
The Julian day number therefore denotes the number of days that has elapsed, at
Greenwich noon on the day designated, since the preceding epoch. The Julian date
(JD) corresponding to any instant is, by a simple extension of this concept, the
56 EXPLANATORY SUPPLEMENT
Julian day number followed by the fraction of the day elapsed since the preceding
noon.
Julian dates can be expressed in Universal Time or dynamical time, though the
timescale should be specified if precision is of concern. A Julian date expressed in
UT will differ by AT from the same date expressed in TDT. It is not necessary to
know the exact difference between dynamical and Universal Time when specifying
the Julian day number in dynamical time. The timescale, when it needs to be
specified, should be given after the Julian date; e.g., JD 2451545.0 TDT.
Prior to 1984, the term Julian ephemeris date (JED) was used to distinguish the
Julian date in ephemeris time, with the day beginning at 12h ET, from the Julian
date in Universal Time, with the day beginning at 12" UT, The Julian ephemeris
date is equivalent to the Julian date in Terrestrial Dynamical Time.
The fundamental epochs of celestial reference coordinate systems are properly
on Barycentric Dynamical Time (TDB). Thus J2000.0 is 2000 January 1.5 TDB,
which is JD 2451545.0 TDB.
In timekeeping laboratories, for many space activities, and for other applica-
tions, a Modified Julian Date is often used. It is defined by MJD = JD - 2400000. 5.
Thus a day of MJD begins at midnight of the civil day. The timescale should be
specified when precision matters, e.g., MJD 47479.25 TAI. Note that MJD is used
in Table 2.58.1 (page 86).
2.27 Time Zones
Universal Time is equivalent to the standard time for 0° longitude, which is defined
to be the meridian through Greenwich, England. A worldwide system of standard
time zones is based on 15° (i.e., one hour) increments in longitude. These standard
time zones are identified with letter designations, in addition to appropriate names.
Thus the standard time zone for 0° longitude is labeled Z. The standard time zone
for 15° east longitude is labeled A. Subsequent letters designate zones increasing
eastward in 15° increments, with M designating the zone for 180° east. The time
zone for west 15° is labeled N, with letters progressing to Ffor 180° west longitude.
In practice, however, each country selects the appropriate time zone, or zones,
for itself. The exact zone boundaries within each country are determined through
that country's internal political process. In addition, some countries disregard the
aforementioned convention by adopting a zone on a fraction of an hour. Many coun-
tries also adopt daylight saving time, sometimes called summer time or advanced
time, which is generally one hour in advance of the standard time normally kept
in that zone. The dates for changing to and from daylight saving time vary among
the countries; a few countries retain advanced time all year.
Figure 2.27.1 is a world map of the different time zones. Since the zones adopted
by individual countries are subject to change, details of the map may become ob-
solete.
Figure 2.27.1
World map of time zones
58 EXPLANATORY SUPPLEMENT
2.3 PRACTICAL DETERMINATIONS OF TIME
Today, International Atomic Time (TAI) represents the most precise achievable
determination of time. This timescale is completely determined by applying the
methods of physics to the determination of time. However, we live and make astro-
nomical observations on the surface of the rotating Earth; thus, it is necessary to
relate TAI to the less consistent timescale determined from the Earth's rotation.
The most recent information on time standards and methods of disseminating
time and frequency is given in the volumes of Proceedings of the Precise Time and
Time Interval (PTTI) Applications and Planning Meeting and the documents of
CCIR Study Group 7. A review of frequency standards was given by Audoin (1984).
2.31 Frequency Standards and Clock Performance
To produce uniform time intervals, a clock must generate a frequency that is accu-
rately known with respect to a standard. The standard frequency for atomic time
was established by Markowitz et al. (1958), who measured the resonance frequency
of Cesium in terms of the ephemeris second. The currently adopted standard is a
liyperfine transition of the Cesium-133 atom (F = 3, MF = 0 to F = 4, MF = 0), which
is assumed to be 9192631770 Hz in a magnetic field of H = 0. A description of the
physics involved and a review of frequency standards are given by Mockler (1961),
Beehler et al. (1965), and Vanier and Audoin (1989).
No clock is perfect. Since all clocks are subject to random-noise rate variations
and systematic effects, a perfect clock is an ideal that can only be approached,
but not reached. Improvements in clock performance require improvements in the
signal-to-noise ratio of the carrier power to phase-side-band power. Given these
facts, it is necessary to be able to specify clock performance.
We first consider a clock as a generator of a nearly sinusoidal signal voltage:
V(t) = [V0 + e(f)] sin $(/). (2.31-1)
For our purposes, e(t) will be very small or will vary slowly and we can ignore it for
most timing applications; t is the ideal time or the time of a reference clock. We
can write for the instantaneous phase
$(r) = Qr + 0(f), (2.31-2)
where
(2.31-3)
2 / TIME 59
is the ideal or reference frequency, £1 the circular frequency in rad/s, and 0(f) the
phase error of the clock. The clock error is then
x(0 = 0(0 /£X (2.31-4)
and the instantaneous normalized frequency error is
dx „ , dS(i)
'"IT <2'31-5>
The average normalized frequency error between f[ and t2 is given by
. x(t2)-x(t]) 1 Z"1
y= — : : = : r / y(t)<*t. (2.31-6)
All measures of clock performance will be related to y(0 or x(t), with age, envi-
ronmental conditions, etc., as qualifying parameters. However, for various reasons,
frequency aspects y(t) have been emphasized over accumulated clock error x(t). The
basic problem in any measure of clock performance is the fact that x{i) as well as
y(0 may be unbounded as t goes to infinity.
For details of characterization of clock performance see Allan (1985), Barnes
(1976), Allan (1983), Jenkins and Watts (1968), and Loeb (1972).
2.3 1 1 'types of Clock Noise Various types of random clock disturbances have been
recognized. Some of these are called white phase noise, white rate noise, and random
walk in rate. Barnes and Allan (1964) recommended measures of clock performance
that do not depend critically on data length or the cut-off frequencies in the per-
formance of drifts. One solution is to take the average of the variances of only two
samples. This statistic is now generally accepted and designated either pair variance
or Allan variance. This is defined by
^) = 4£Cyr-»-i)2, (2.311-1)
k=]
where yk and yk-\ represent successive samples y computed as in Equation 2.31-6,
but with the conditions that t2 - 1\ = t be the same for all samples and that there
is no dead time between samples.
Allan (1966) investigated the dependence of the general sample variance on the
parameters of dead time, integration time, number of measurements in the sample,
and measurement-system bandwidth. Further details are found in Barnes et al.
(1971), Howe (1976), and CCIR Report No. 580.
gO EXPLANATORY SUPPLEMENT
2.312 Clock Modeling The Allan variance (Equation 2.311-1) is a function of r.
Starting with sufficiently small values of r (a fraction of a second) and letting t
increase, we observe that, for all clocks, o{t) is first decreasing (frequency white
noise), then reaches a smooth minimum (frequency flicker-noise floor), then in-
creases (frequency random walk). The values of r corresponding to the flicker- noise
floor indicate the ability of the clock to ensure short-term or long-term stability.
The knowledge of ct(t) versus r is essential for deciding which clock should be used
in applications. It is also necessary to recognize abnormal behavior (Percival, 1976)
in order to predict the clock frequency and to maintain synchronizations. Methods
of optimal prediction have been discussed by Barnes (1976). Further references are
Box and Jenkins (1970) and Anderson (1975).
The methods of characterizing clock frequency can also be applied to computed
timescales, such as TAI, under the assumption that such timescales are the output
of fictitious clocks.
2.32 Measurement of Atomic Time
2.321 Quartz-Crystal Oscillators Although they measure molecular rather than
atomic resonances, quartz-crystal oscillators play a fundamental role in timekeeping.
They are almost indispensable sources of stable signals in all systems, including
atomic clocks. They can be free-running or locked to a reference signal, with time
constants ranging from milliseconds to days. The performance characteristics may
range from 10-4 to 10~13 in frequency stability per day. Gerber and Sykes (1966)
give a review of quartz-crystal oscillators.
2.322 Cesium Beam Standards Cesium-beam frequency standards can be divided
into two categories. One is laboratory cesium-beam frequency standards. These re-
alize the second with the utmost accuracy (currently, 1.5 X 10-14) and are, there-
fore, stable in the very long term. As of 1988, operating instruments are located
in Canada, the Federal Republic of Germany, Japan, the United States, and the
USSR. Other instruments are under construction in several countries. The other
category includes commercially available frequency standards. These devices are
less accurate (2 X 10~12), but may equal the laboratory standards for stability up
to about one year. The typical stability of such high-performance clocks is 3 x 10-'4
for a time interval of one day (Percival, 1973).
2.323 Hydrogen Masers The maser is the instrument of choice for applications
where the utmost stability is needed for integration times from ten seconds to ten
days, regardless of cost. The optimum stability reaches about 1 x 10-15 for integra-
tion times of 1000 to 10000 seconds. The clock transition is the transition P - 1,
MF = 0 to F = 0, MF = 0, with a transition frequency /h - 142040575 1 . 768 Hz (Rein-
hardt and Lavanceau, 1974). The frequency of transition is dependent on the cavity
2 I TIME
61
size and temperature (Peters and Washburn, 1984). Nevertheless, some hydrogen
masers equipped with automatic tuning of the cavity and kept in temperature-
controlled rooms have a long-term stability of the same order as the best cesium
standards, Ramsey (1968) gives principles and scientific background for masers that
make it clear why the hydrogen maser has gained such high interest among physi-
cists. Kleppner (1965) gives a discussion of the various physical effects governing
the maser's behavior, along with detailed technical considerations.
2.324 Rubidium Vapor Cells The rubidium clock is an appropriate device when
a relatively low-cost clock is needed that has better stability than a quartz-crystal
clock. The rubidium clock can reach a stability of 1 x 10~13 per day under the
best conditions, but it is subject to temperature- and pressure-induced frequency
variations. Ringer et al. (1975) describe the design and performance of a clock for
the GPS satellites. This clock has achieved stability of 2 x 10-13 per day.
2.325 Mercury Ion Frequency Standard The mercury- ion frequency standard uses
ions that are confined in a small region of space by an electromagnetic field trap.
Thus the particles can be observed without having them collide with the walls,
which would disturb the atomic resonance. The mercury isotope Hg-199 has an
extremely narrow microwave resonance line at 40507 MHz. Although this type of
frequency standard should be a large improvement over the cesium standard, it has
an unfortunately low signal-to-noise ratio, resulting in limited short-term stability.
However, the long-term stability is very good, since integration can take place over
a number of days (Winkler, 1987).
2.33 Earth Rotation Measurement
At the present time accurate determinations of UTl and polar motion are based on
VLBI, satellite laser-ranging, and lunar laser-ranging data. The different techniques
are shown in Table 2.33.1.
2.331 Optical methods Universal Time can be determined from transit instru-
ments, in particular the Danjon astrolabe and the photographic zenith tube (PZT)
(Markowitz, 1960). The methods used are those of classical spherical astronomy
as presented in textbooks (Mueller, 1969; Woolard and Clemence, 1966; Danjon,
1959). Optical observations can determine UT to about 5 ms of time.
The classical optical methods are restricted by three major problems: atmo-
spheric refraction, star-catalog errors, and spurious deflections of the vertical. Cur-
rent estimates of the capabilities of the various methods for Universal Time and
polar motion are given in Table 2.33.1. As there are generally correlations between
successive measurement errors, increasing the averaging time by a factor / does not
result in a decrease of the uncertainties by f~m.
62
EXPLANATORY SUPPLEMENT
Table 2.33.1
Techniques for Measuring Earth's Rotation
Technique
Polar
Universal
Averaging
Measurement
Motion {")
Time (ms)
Time
Conditions
Astrolabe
0.06
4
2 hours
one instrument
Photographic Zenith
0.04
4
2 hours
one instrument
Tube
Optical Method
0.01
0.8
5 days
network of 80 instruments
Doppler
0.01
2 days
1 satellite, 20 stations
L.LI
n ni
1
1
3 days
VLBI
0.001
0.05
24 hours
network of 5 or 6 stations
0.1
1 hour
2 stations
Satellite Laser-Ranging
0.001
0.1
3 days
network, - 20 stations
(LAGEOS)
Lunar Laser-Ranging
0.1
1 day
1 instrument*
* Under good meteorological conditions
2.332 The Doppler Method The Doppler satellite system uses as a reference a
network of stations distributed over the globe. The system does not have a fixed
reference in space and is dependent upon the dynamics of the satellites and the
knowledge of the positions of the Earth-based stations. A report on this method
is given by Anderle (1976). The Doppler method provides length of day and polar
motion values, but cannot determine Universal Time.
2.333 Lunar Laser-Ranging Laser-pulse ranging between Earth-based telescopes
and retroreflectors installed on the surface of the Moon is currently done to a preci-
sion of approximately 2 cm. A capability of determining Universal Time to a fraction
of a millisecond was demonstrated by Silverberg (1974) and is effectively shown by
recent results (Newhall et ai, 1988; Veillet and Pham Van, 1988).
Because of the precision of the measurements, these observations are very sen-
sitive to the dynamics of the Earth-Moon system. They provide very accurate
corrections to the theory of the Moon's motion and librations and the theory of
general relativity. Other significant parameters are those that affect the positions
of the Moon-based retroreflectors and the Earth-based telescope — including Earth
tides, continental drift, plate motion, and polar motion. The observations are also
sensitive to the constant of precession, the theory of nutation, and the Earth's orbit.
All of these parameters are correlated in complex ways.
2.334 Radio Interferometry The rotational position of the Earth is measured with
respect to a plane parallel wave from a cosmic radio source that is so far away that
it can be safely assumed to be without proper motion. Very Long Baseline Inter-
2 / TIME
63
feromefry (VLB!) and Connected Element Interferometry (CEI) can be used to
determine Universal Time and polar motion with excellent resolution. VLBI of-
fers higher precision. In addition, it provides synchronization between clocks at the
participating observatories to a precision of a fraction of a nanosecond. But since
the observations from participating observatories must be correlated, VLBI has the
disadvantage of a delay between the time of observation and the availability of the
results. CEI techniques have the advantage of being less expensive and providing
immediate results. VLBI techniques are described by Counselman (1976) and Meeks
(1976). Short baseline techniques are described by Johnston (1974).
2.34 Dynamical Time Determinations
The IAU (1976) definition of dynamical time (see Section 2.22) relates the dynam-
ical time of solar-system ephemerides to the timescale of terrestrial atomic physics.
However, distinctions must be made between an ideal dynamical time, which is
the independent variable in the equations of motion for the solar system, and both
TDT and TDB, which are currently defined as ideal forms of atomic time.
In principle, the time in the equations of motion of the Sun, Moon, and planets
could diverge from the time determined from observations of phenomena of terres-
trial physics. At the present time, observational determinations are not sufficiently
accurate to indicate such a systematic difference. If a true difference were detected,
the scientific community would have to decide how to accommodate that difference.
An ideal dynamical time can be determined only by analysis of observations over
an extended period of time.
Occultation observations, specifically observations of the time when the Moon
passes in front of a star, provide the most accurate current method for investigating
dynamical time. The observations require a recording of the time when the occul-
tation took place and the location of the observer. The predicted dynamical time
of the occultation can be compared with the recorded atomic time of the observed
phenomenon to provide information concerning the difference between dynamical
time and International Atomic Time (TAI).
A difference between dynamical time and TAI could also be revealed if different
methods of determining Universal Time give discordant results. In particular, a
disagreement between Lunar Laser-Ranging and Very Long Baseline Interferometry
could indicate nonmodeled variations of the Moon's right ascension. This would
provide a determination of dynamical time.
Future analysis of high-precision planetary observations could reveal a differ-
ence in the timescales if differences between predicted and observed positions are
systematically proportional to the mean motions of the planets. At the present
time, the planetary residuals do not indicate such a systematic deviation, and the
possibility is only a conjecture.
64
2.4 METHODS OF TIME TRANSFER
EXPLANATORY SUPPLEMENT
The determination and maintenance of a timescale at its source is a distinct prob-
lem from the dissemination of that timescale to many sites around the world. The
appropriate method of disseminating the timescale is largely dependent upon the
accuracy required. Also, the appropriate method can be expected to change with
time as new and improved technologies become available.
Although this section will discuss the various techniques currently in use and
the accuracies achievable, it can be expected that the contents of this section will be
one of the first to be out of date. The most up-to-date information on this subject
can be found in the documents of the Consultative Committee on International
Radio (CCIR) Study Group 7: Standard Frequencies and Time Signals. A review
is given by Klepczynski (1984). A CCIR Handbook on satellite time and frequency
dissemination is in preparation.
There are two main techniques for transferring time: transportation of an op-
erating clock and sending of time signals. The latter technique subdivides into
one-way and two-way methods.
In the one-way methods, knowledge of the propagation delay is required and is
often a limitation to the accuracy. For instance, the traditional high-frequency time
signals are subject to delay variations and an accuracy of only 1 ms can be achieved.
But in the use of the Global Positioning System (GPS) satellites, the propagation
can be well-modeled, and determination of time from a specific satellite has an
uncertainty in the 10-20 ns range.
One mode of operation of the one-way technique is to receive the same sig-
nal at two or more sites. After the exchange of data, the difference of readings of
the clocks is obtained at the emitter. Only the differential delay of propagation
is needed. This method allows for accurate time comparisons using the signals of
commercial television, LORAN-C, GPS and geosynchronous satellites, and natural
astronomical sources such as VLBI.
Two-way techniques involve the nearly simultaneous exchange of time signals
back and forth, along the same path. The uncertainties are reduced because the
propagation delay is eliminated in the data processing. Usually the method is ap-
plied using a satellite to relay the signal. This requires complex equipment. Laser
pulses can be used either for a terrestrial link or for a link involving reflection on
an artificial satellite. In the USSR, the two-way technique is routinely used with re-
flection on meteor trails, with accuracies reaching 20-25 ns. When the uncertainties
of the time transfers are smaller than 1 /is, relativistic effects must be considered,
as explained in Section 2.47.
2 / TIME
2.41 Radio Time Signals
65
Radio time signals continue to be a simple means of distributing time. The trans-
mitted time is UTC, with a code or voice transmission of DUT1 so that UTl can
be obtained using the relation UTl - UTC + DUT1. A number of countries have
standard-frequency transmissions with a wide coverage and an extended period
of continuous transmission. The common frequencies for transmission are 2.5, 5,
10, 15, 20, and 25 Mc/s. However, a number of countries also have transmissions
at special frequencies and specific times. Report 267-5 of CCIR. Study Group 7
and the Annual Report of the time section of BIPM give the characteristics of
standard-frequency and time-signal emissions in allocated bands, and characteris-
tics of stations emitting with regular schedules with stabilized frequencies outside
of allocated bands.
The accuracy that can be obtained is severely restricted by unknown variations
in the travel time of the radio signal, particularly on the higher frequencies. The
received signal may contain, in varying quantity, signals that have traversed differ-
ent transmission paths. Reception conditions may differ widely between different
reception sites and, at any one site, there may be a considerable diurnal variation.
The best result is generally achieved by measuring transmissions at different times
throughout the day and by taking suitably weighted means. Anomalies are partic-
ularly troublesome when sunrise or sunset occurs on the transmission path; these
times should be avoided if possible. The most serious discordances may occur when
the reception site is within the skip area of the transmitter, since the predominating
signal may be received by backscatter from a distant point, either on the ground
or in the ionosphere.
2.42 Portable Clocks
The use of portable atomic clocks for the comparison of timescales has been the
most reliable and accurate method. It has also been one of the most expensive
techniques on a cost-per-comparison basis. Portable clocks have been necessary for
calibrating path delays for synchronization by other means. Portable atomic clocks
were first used by Reder and Winkler (1960) in experiments conducted in 1959 and
1960. These early atomicrons were 7-feet high and weighed several hundred pounds.
They were portable in the sense that they could be transported while running, but
they required extensive external power. In the early sixties, the size of cesium-beam
clocks was reduced enough that two men could, with difficulty, transport a clock.
This was demonstrated by Bagley and Cutler (1964).
66 EXPLANATORY SUPPLEMENT
In 1965 the U.S. Naval Observatory established a portable clock service, which
has provided thousands of clock synchronizations. The portable clock consists of a
commercial cesium-beam oscillator, a counter or clock, and a power supply with
rechargeable batteries, which are capable of approximately eight hours of continu-
ous operation without connection to alternating current. The clock system weighs
approximately 68 kg and measures approximately 0.4 m by 0.4 m by 0.53 m. The
usual precautions of minimizing shock, vibration, heat, cold, and magnetic fields
are mandatory to yield high accuracy on portable clock trips.
For a clock trip, the portable atomic clocks are adjusted and kept as close to
the frequency of the master clock as possible. The time is set so that the difference
between the portable clock and the master clock passes through zero at the mid-
point of the trip. On the day of departure, clock measurements are made against
the reference clock systems of the master clock with a resolution better than a
nanosecond, and with a small, portable, ten-nanosecond counter that is taken on
the trip. During the trip, clock data are recorded for each individual clock. When
the clock is returned, it is again compared with the master clock. Once post-trip
data are available, both the frequency offset of the portable clock and the reduction
and analysis of the trip data can be completed. For accurate results, the changes in
the gravitational fields on the clock and the clock velocity relative to ground-based
or stationary clocks must be considered (see Section 2.47).
2.43 LORAN-C
The LORAN-C navigation system has been routinely used as a time-transfer stan-
dard for time and frequency coordination since 1969. The LORAN-C system con-
sists of chains, each of which has a master transmitter and two to four secondary, or
slave, transmitters. Each of the master transmitters is separated by several hundred
kilometers. The master transmitter emits a series of coded 100-kHz bursts that axe
repeated after specified and controlled delays by the secondary stations. Bursts of
100-kHz transmissions enable the receiver to separate the stable ground- wave signal
from the sky-wave signal that arrives 40 ms or more later. A navigation or timing
receiver will track the third cycle of the received bursts. Each LORAN-C station
has a cesium standard for deriving time signals. The complete chain is overseen by
one or more monitor stations that measure and control the emission delays. Navi-
gation accuracy within the ground-wave coverage of the chain is 0.10-0.25 nautical
miles. Ground-wave coverage of the LORAN C transmitter is highly variable, and
is a function of the radiated power (275-1800 kilowatts) and the conductivity of
the path. Over land, ground- wave coverage is typically 500-1000 km; over water
this coverage can be over 2000 km. Since each LORAN-C chain is independent,
time comparisons with LORAN-C are restricted to laboratories that have com-
mon ground-wave views of a single chain. With a single chain, the monitor stations
maintain self-consistent timing to better than a microsecond. The uncertainty of
2 / TIME
67
a LORAN-C time comparison is on the order of 50-100 ns for one day, with the
uncertainty increasing to 300 1000 ns for one hundred days. This increase is due
to annual weather-related effects over the LORAN-C ground-wave path. Changes
in ground conductivity as well as temperature will change the path delay of the
LORAN-C ground-wave signal. Uncertainty of the path delay is the physical lim-
itation of LORAN-C timing networks, and is independent of the quality of the
LORAN-C system or the LORAN-C receiver.
2.44 Television Comparison Techniques
Another technique for time or frequency comparisons is television (TV) transmis-
sions or high-power, wide-bandwidth communication links that broadcast in VHF
and UHF frequency bands. (These bands have small propagation delay variations.)
Television receivers are inexpensive and readily available. There are two basic types
of TV transfer links. The first is the differential transfer link, in which both clocks
are in common view of a single TV transmitter. The second includes a microwave-
network link as part of the differential path. The common-view transfer link is
much more stable, since only the different propagation paths through the atmo-
sphere affect the measurement. Long-term time-transfer stability of 10-150 ns can
be expected. Such transfer links have been used for comparison for everything from
remote clocks to standard timing laboratories.
A clock-comparison link that includes a network microwave system will have
greater time-transfer uncertainty. The differential path includes delays from user
to transmitter, delays from studio to transmitter link, delays from an intermediate
studio, delays through multiple microwave repeaters to the network origination
point, and delays from the transmitter to the other user. Except for the delay
between the user and the transmitter, the delays are subject to step changes from
a hundred nanoseconds to several-hundred microseconds.
To ensure that the TV programs remain on the air, the links are redundant
for reliability. Therefore, if there is a bad studio-to-transmitter link, a backup link
will be switched in its place. These changes to the backup can cause differences
in the delay of as much as 500 nanoseconds. Due to this redundant characteristic
of the TV networks, the networks are largely unsuitable for high-accuracy time
coordination.
Another factor that reduces the use of TV-network transfer links is the use
of digital frame synchronizers by the network stations. The purpose of the frame
synchronizer is to synchronize the network picture with all locally generated signals,
thus preventing picture rolls when switching between a network program and a local
program. A digital frame synchronizer continuously digitizes and stores an incoming
TV signal. The stored digital information is continuously read out and converted
back to an analog video signal under control of the station's master synch generator.
The synchronizing pulses for the network picture have therefore been replaced by
68 EXPLANATORY SUPPLEMENT
local station synchronizing pulses. Thus network synchronizing pulses that are used
for the time transfer have been lost.
2.45 Use of Satellites
Both the Transit satellite system and the Global Positioning System (GPS) can
be used for one-way time transfer. Each Transit satellite continuously transmits its
current ephemeris encoded by a phase modulation on two stable carrier frequencies
of approximately 150 MHz and 400 MHz. The navigational information is broadcast
in two-minute intervals, which begin and end at the instant of each even minute.
An encoded time marker is broadcast, with time uniquely marked at the instant of
the even minute.
The broadcast frequencies are based on the highly precise master oscillator.
The Transit satellite system is expected to be replaced by the NAVSTAR GPS
navigation system. This system calls for a constellation of 18 satellites that are
distributed in orbital planes inclined at 63° with respect to the Earth's equatorial
plane. Although the GPS system is designed for navigation capabilities, it also
coordinates most of the major timescales in the world. The GPS system has an
accuracy of about 10 ns for ten-day averaging. The most accurate time comparison
by GPS is based on a common satellite procedure. GPS satellites on a flyby mode
make two separate comparisons with respect to the GPS satellite clock. A linear rate
for the satellite clock is assumed. This method does not provide the same accuracy
as the common-view procedure, since the clock error and ionospheric errors will
be present and more significant. The error budget for a common-view GPS time
transfer is given in Table 2.45.1.
Table 2.45.1
Error Budgets for Common View GPS Time Transfer
Type of Delay
Best Case
Worst Case
RMS
RMS
Satellite Ephemeris
3 ns
10 ns
Ionospheric
2 ns
100 ns
Tropospheric
1 ns
20 ns
User-Position
1 ns
1 ns
Multipath
1 ns
2 ns
Receiver
1 ns
1 ns
Signal-To-Noise
7 ns
1 ns
Total RMS
4.2 ns
103 ns
(for single 13-minute track)
Note: Some of the errors depend on the distance (for example the
satellite ephemeris). The best case applies to distance of
2000-3000 km; the worst case is for distance of 6000-8000 km.
2 / TIME
69
With one-way time transfer, the accuracy of the method is basically limited
to the accuracy of the ephemeris or the knowledge of the satellite's position. In a
two-way time transfer, a time-interval counter is located at each Earth station. The
counter is started by a pulse from a local clock and stopped by a pulse received via
the satellite from the remote clock. Thus the time interval between pulses involves
the transmitter delay, the propagation delay between the Earth station and the
satellite, the delay in the satellite for the signal received, the propagation delay
between the satellite and the other Earth station, and the same list of delays on
the return direction. In addition, the effect of the Earth's rotation must be taken
into account, because during the period of the time transfer, both Earth stations
have moved and the satellite has moved. Thus each leg of the comparison must
be considered individually, and they cannot be assumed to be equivalent to the
corresponding leg at a different time. The differential delay effects caused by the
Earth's atmosphere and the different frequencies of the different legs should not
exceed a few nanoseconds, whether due to geometric effects or to velocity effects
caused by the ionosphere and troposphere. At frequencies of 20-30 GHz, the effects
of humidity may cause significant inequalities for time transfers at the nanosecond
level (Oaks, 1985).
2.46 Intercontinental Clock Synchronization by VLBI
With Very Long Baseline Interferometry (VLBI), radio signals from celestial objects
are recorded at individual remote radio telescopes. The observational data are then
brought together for correlation so that precise values for the differences between
the clocks at the individual antennas can be determined. The fundamentals of
VLBI have been described by Klemperer (1972), who also presents an extensive list
of references. Several VLBI experimental programs, and the fundamentals of the
process, were described by Clark (1972). Frequency comparisons with an accuracy of
10-13 to 10-14 and clock synchronization of the order of 1 ns are possible. To achieve
this accuracy, 110-megahertz bandwidths are required. These can be achieved by the
bandwidth synthesis techniques described by Hinteregger et ai (1972) and Rogers
(1970). The VLBI process requires distinct measurements based on prearranged
experiments involving different locations. The results of the time comparisons are
not available until the two data sets are brought together and correlated. In theory,
if sufficient data are available and a sufficiently large bandwidth can be obtained,
the VLBI synchronization method is mainly limited in accuracy by the difficulty
of determining the overall system delay and the atmospheric or ionospheric delay
(Johnston et ai, 1983).
70
2.47 Relativistic Effects in Time Transfer
EXPLANATORY SUPPLEMENT
When transferring time between two points, the process can be viewed from one
of two reference frames — a geocentric, Earth-fixed, rotating frame or a geocentric,
nonrotating, local inertial frame. Discussions of time transfer viewed from these
two frames are given in the following sections. The equations used in the discus-
sions represent clock rates with an accuracy better than 1 in 1014. In addition, the
equations are consistent with the proposal of the Consultative Committee for the
Definition of the Second (see Section 2.211), but they extend the procedures to
heights that include geostationary satellite orbits.
2.471 Clock Transport from a Rotating Reference Frame When time is transferred
from point P to point Q by means of a portable clock, the coordinate time accu-
mulated during transport is
At = J d
AU(r) v2
+ -^-AE, (2.471-1)
where c is the speed of light; w is the angular velocity of the Earth's rotation; v is
the velocity of the clock with respect to the ground; r is a vector from the center of
the Earth to the clock, which is moving from P to Q\ Ag is the equatorial projection
of the area swept out during the time transfer by the vector r as its terminus
moves from P to Q; AU(r) is the potential difference between the location of the
clock at r and the geoid as viewed from an Earth-fixed coordinate system (with
the convention that At/(r) is positive when the clock is above the geoid); and ds is
the increment of proper time accumulated on the portable clock. The increment of
proper time is the time accumulated on the portable standard clock as measured
in the rest frame of the clock; i.e., the reference frame traveling with the clock. Ae
is measured in an Earth-fixed coordinate system. As the area Ae is swept out, it is
taken as positive when the projection of the clock's path on the equatorial plane
is eastward. When the height h of the clock is less than 24 km above the geoid,
AU{r) may be approximated by gh, where g is the total acceleration due to gravity
(including the rotational acceleration of the Earth) evaluated at the geoid. When h
is greater than 24 km, the potential difference AU{r) must be calculated to greater
accuracy as follows:
AV(r) = ~GMe(1--1-)- ioV sin2 9-a\)+ ^ [l + ( *Y (3cos2 0-1)1.
(2.471-2)
In this equation, at is the equatorial radius of the Earth; r is the magnitude of
the vector r; 8 is the colatitude; GMe is the product of the Earth's mass and the
gravitational constant; and J2 is the quadrupole moment coefficient of the Earth
{J2 =+1.083 x 10~3).
2 / TIME
71
2.472 Nonrotating, Local Inertial Reference Frame When time is transferred from
point P to point Q by means of a clock, the coordinate time elapsed during the mo-
tion of the clock is
At
is
1 -
U(r) - t/g v2
72 + 2?
(2.472-1)
Here E/(r) is the potential at the location of the clock and v is the velocity of
the clock, both as viewed — in contrast to Equation 2.471-1 — from a geocentric
nonrotating reference frame. Ug is the potential at the geoid, including the effect
on the potential of the Earth's rotational motion. Note that AU{v) ^ U(r) - Ugl
since U(r) does not include the effect of the Earth's rotation. This equation also
applies to clocks in geostationary orbits, but should not be used beyond a distance
of about 50000 km from the center of the Earth.
2.473 Electromagnetic Signals Transfer from a Rotating Reference Frame From
the viewpoint of a geocentric, Earth-fixed, rotating frame, the coordinate time
elapsed between emission and reception of an electromagnetic signal is
c Jp L c
where da is the increment of standard length, or proper length, along the transmis-
sion path; and AU(t) is the potential at the point r on the transmission path less
the potential at the geoid (see Equation 2.472-1), as viewed from an Earth- fixed
coordinate system. Ae is the area circumscribed by the equatorial projection of the
triangle whose vertices are at the center of the Earth; at the point of transmission
of the signal, P; and at the point of reception of the signal, Q.
The area Ag is positive when the signal path has an eastward component . The
second term amounts to about a nanosecond for a round-trip trajectory from the
Earth to a geostationary satellite. In the third term, 2u>/c2 = 1. 6227 x 10-6 ns km-2;
this term can contribute hundreds of nanoseconds for practical values of AE. The
increment of proper length da can be taken as the length measured using standard
rigid rods at rest in the rotating system. This is equivalent to measurement of length
by taking c/2 times the time (normalized to vacuum) of a two-way electromagnetic
signal sent from P to Q and back along the transmission path.
2lo
rAE,
(2.473-1)
2.474 Electromagnetic Signal Transfer from a Nonrotating, Local Inertial Frame
From the viewpoint of a geocentric nonrotating, local inertial frame, the coordinate
time elapsed between emission and reception of an electromagnetic signal is
Ar
.if
c Jp
da
1 -
U(r)-UE~
(2.474-1)
72 EXPLANATORY SUPPLEMENT
where U(t) and Ug are defined as in Equation 2.472-1, and da is the increment of
standard length, or proper length, along the transmission path. The quantities da
appearing in Equations 2.473-1 and 2.474-1 differ slightly because the reference
frames in which they are measured are rotating with respect to each other.
Example 1 Due to relativistic effects, a clock at an elevated location will appear to be
higher in frequency and will differ in normalized rate from TAI by
AUt/c1, (2.474-2)
where AUt is the difference in the total potential (gravitational and centrifugal potentials),
and where c is the velocity of light. Near sea level this is given by
g(<j>)h I c1, (2.474-3)
where g(4>) = (9. 780 + 0, 052 sin2 0) m/s2 is the total acceleration at sea level (gravita-
tional and centrifugal), <j> is the geographical latitude, and h is distance above sea level.
Equation 2.474-3 must be used in comparing primary sources of the Sf second with TAI
and with each other. For example, at latitude 40°, the rate of a clock will change by
+ 1.091 x 10-13 for each kilometer above sea level.
Example 2 If a clock is moving relative to the Earth's surface with the speed v, which
may have the component v- g in the direction to the east, the normalized difference of the
frequency of the moving clock from that of a clock at rest at sea level is
1 f + gi^h _ 1 wr(cQS ^)VE; (2.474^1)
2 cl cl cl
where u> is the angular rotational velocity of the Earth [uj = 7.992 X 10-5 rad/s), r is the
distance of the clock from the center of the Earth (r = 6378. 140 km), c is the velocity of
light (c = 2.99792458 x 105 km/s), and 4> is the geographical latitude.
Example 3 A clock is moving 270 m/s eastward at 40° latitude at an attitude of 9 km.
The normalized difference of frequency of the moving clock relative to that of a clock at
rest at sea level is
-4.06x 10~13 +9.82x 10~13 - 1.071 x 10-12 =4.95 x 10"13. (2.474-5)
The choice of a coordinate frame is purely a discretionary one, but to define coordinate
time, a specific choice must be made. It is recommended that for terrestrial use, a topocen-
tric frame be chosen. In this frame, when a clock B is synchronized with a clock A by a
radio signal traveling from A to B {both clocks being stationary on the Earth), these two
clocks differ in coordinate time by
B-A = % / r2 cos2 (MA,
c- Jp
(2.474-6)
2 / TIME
73
where <f> is the latitude, A is the longitude (the positive sense being eastward), and P is the
path over which the radio signal travels from A to B. If the two clocks are synchronized
by a portable clock, they will differ in coordinate time by:
r1 cos2 4>d\, (2.474-7)
where v is the portable clock's ground speed, and P is the portable clock's path from A
to B.
This difference can be as much as several tenths of a microsecond. It is recommended
that Equations 2.474-6 and 2.474-7 be used as correction equations for long-distance
clock synchronization. Since these equations are path dependent, they must be taken into
account in any self-consistent coordinate time system.
If a clock is transported from a point A to a point B and is brought back to A on a different
path at infinitely low speed at h = 0, its time will differ from that of a clock remaining in
A by
Ai = -2u}AE I c2, (2.474-8)
where Ae is the area denned by the projection of the round trip path onto the plane of the
Earth's equator. Ae is considered positive if the path is traversed in the clockwise sense
viewed from the South Pole.
Example 4 Since
lu/c1 = 1.6227 x 10~6nskm~2, (2.474-9)
the time of a clock carried eastward around the Earth at infinitely low speed at h = 0 at
the equator will differ from a clock remaining at rest by —207.4 ns.
2.5 HISTORICAL DEVELOPMENT OF TIMEKEEPING
2.51 Introduction
The earliest attempts to measure time were probably based on observations of ce-
lestial and meteorological phenomena: the seasonal motion of the Sun along the
horizon, the progression of Moon phases, the shifting of winds, and so on. Time-
keeping, which we take to mean the measurement of subdivisions of the day, has
long involved observations of both celestial and terrestrial phenomena. On the ce-
lestial side is the measurement of shadows cast by a vertical gnomon, which led
to the development of sundials. On the terrestrial side is the development of water
clocks {clepsydrae). The origins of both approaches predate historical records. His-
torical references to water clocks are found in Babylonian and Egyptian records of
the sixteenth century B.C. (Turner, 1984).
74 EXPLANATORY SUPPLEMENT
The system of hours developed first in Egypt, where day and night were each
divided into 12 parts. This resulted in a system in which the lengths of daylight
and nighttime hours varied with the seasons. In summer, an hour of daylight was
longer than an hour of darkness; in winter, the situation was reversed. Only at
the equinoxes were hours of day and night of equal length. When this system of
temporal hours spread throughout Europe, it became a function of latitude as well
as season. It continued in civil use until the spread of mechanical clocks in the
fifteenth and sixteenth centuries encouraged the gradual acceptance of equal hours.
Our familiar system of 24 equal hours comes from Greek astronomy. At least
as early as Hipparchus, the Egyptian system of 24 temporal hours was merged with
the Babylonian division of the day into 360 equal units. The result was a system of
24 equal hours, with each hour divided into 60 minutes (Neugebauer, 1975). This
established a tradition for astronomical tables that continues to this day.
Although their origin is a mystery, mechanical clocks were in use in Europe
near the beginning of the fourteenth century. They were driven by weights and
controlled with the recently invented escapement. A regulating device allowed the
clocks to be adjusted, morning and evening, to maintain temporal hours. They did
not provide adequate standards for scientific timekeeping, but were built for public
use.
In the mid-seventeenth century, the pendulum was successfully applied to clocks.
By the late eighteenth century, clocks were improved to provide accurate time for
scientific purposes and for navigation at sea.
Useful surveys of the history of timekeeping are provided by Andrewes and
Atwood (1983), Landes (1983), and Whitrow (1988).
2.52 Apparent Solar Time, Mean Solar Time, and the Equation of Time
Apparent solar time is the measure of time defined by the actual diurnal motion of
the Sun. If the Sun's right ascension increased at a uniform rate, solar days would
be equal in length throughout the year (assuming uniform rotation of the Earth).
Apparent solar time would then provide a uniform measure of time. However, the
Sun's motion in right ascension is not uniform because the Sun moves in the ecliptic
rather than the equator, and the Sun's motion in the ecliptic is not uniform.
Mean solar time was defined by the motion of an abstract fiducial point, which
came to be known as the fictitious mean sun. This point was posited to move
uniformly in the equatorial plane at a rate that is virtually equal to the mean rate
of the true Sun's motion in the ecliptic. The difference between mean solar time
and apparent solar time is called the equation of time. It is never larger than 16
minutes.
The concept of mean solar time, together with the principles for determining the
equation of time, extend back at least to Ptolemy. For Ptolemy and his successors,
mean solar time was useful for constructing tables of solar motion. But as long as
2 / TIME 75
the only means of obtaining accurate time was by direct astronomical observation,
apparent solar time was in general use. Determinations of local apparent time were
commonly made by observing altitudes of the Sun or stars. Thus apparent solar time
was the argument in The Nautical Almanac and other national ephemerides until
the early nineteenth century. Mean time, when needed, was obtained by applying
the equation of time to the apparent time.
The equation of time, in the sense of the correction to be applied to apparent
time in order to obtain mean time, was tabulated in the national ephemerides
from their earliest inception. It was used for regulating clocks and determining the
argument for entering astronomical tables. During the late eighteenth and early
nineteenth centuries, as clocks were improved and came into extensive use at sea,
apparent time was gradually superseded in civil use by mean solar time.
In the mid-nineteenth century, when mean time was first introduced as the
argument in the national ephemerides, the equation of time was supplemented by
an ephemeris of sidereal time at mean noon. This provided a means of calculating
mean solar time from sidereal time, without using the equation of time. Clocks were
not sufficiently perfected for the theoretical distinction between apparent sidereal
time and mean sidereal time to be of any practical importance. This distinction was
therefore disregarded, and the very imperfect expressions then in use for nutation
were of no consequence for the purpose of time determination.
The Riefler clock, introduced about 1890, was the earliest timepiece with an
accuracy comparable to determinations of time from observation. As the accuracy
of clocks increased, the explicit recognition of mean sidereal time as distinguished
from apparent sidereal time became necessary. The term uniform sidereal time was
often used at first, but this measure was not strictly uniform, and the term was
dropped from use.
After the introduction of the Short free-pendulum clock in 1921, the removal
of the short-period terms of nutation from the observed clock corrections was nec-
essary in order to check the clock satisfactorily. These terms were included in the
ephemerides of the sidereal time at 0h, beginning with 1933.
The equation of time came to signify the opposite of the original concept.
Apparent solar time was obtained by applying the equation of time to the mean
time kept by clocks, which were regulated by determinations of mean time from
observations of sidereal time. The link between mean solar time and sidereal time
was founded on the relation
mean solar time at Greenwich = GHAQ — Rq + 12h, (2.52-1)
where GHAQ is the Greenwich hour angle of the mean equinox of date (i.e., Green-
wich mean sidereal time) and RQ is the right ascension of the fictitious mean Sun.
Underlying the concept of mean solar time was the assumption that the rotation
of the Earth was uniform. In the first half of the twentieth century, this assumption
76 EXPLANATORY SUPPLEMENT
was repudiated (see Section 2.53), with the result that mean solar time was no longer
used in precise timekeeping. It was replaced by two concepts: ephemeris time was
introduced to satisfy the desire for a uniform measure of time, and Universal Time
(originally introduced to specify Greenwich mean time measured from midnight
instead of noon — see Section. 2.521) came to designate a measure of the Earth's
rotation.
Over the centuries, the exact definition of mean solar time, and hence of the
equation of time, evolved as new tables for the motion of the Sun were introduced.
In analyzing historical observations that are expressed in mean time, care must be
taken to determine what definition was used in the reductions.
2.521 Greenwich Mean Time Prior to 1925, mean solar time was reckoned from
noon in astronomical practice. The mean solar day beginning at noon, 12" after the
midnight at the beginning of the same civil date, was known as the astronomical
day. Mean solar time reckoned from mean noon on the meridian of Greenwich was
designated Greenwich Mean Time (GMT). Local mean time (LMT) was reckoned
from mean noon on a local meridian. Beginning with the national ephemerides for
1925, a discontinuity of 12h was introduced in the tabular arguments, so the instant
designated December 31.5 in the volumes for 1924 was designated January 1.0 in
the volumes for 1925.
In The Nautical Almanac, the designation Greenwich Mean Time (GMT) was
still used for the new reckoning, together with local mean time where appropriate,
whereas in The American Ephemeris the designation Greenwich Civil Time (GCT)
was adopted, together with local civil time (LCT). This confusion in terminology
was finally removed by dropping both designations and substituting Universal Time
(UT). Care is necessary to avoid confusion. To distinguish between the two reckon-
ings that have both been called Greenwich Mean Time, the designation Greenwich
Mean Astronomical Time (GMAT) should be used for the reckoning from noon.
The designation UT always refers to time reckoned from Greenwich midnight, even
for epochs before 1925.
In addition, it should be noted that in the United Kingdom, Greenwich Mean
Time has meant the civil time or Coordinated Universal Time (UTC). For naviga-
tion, Greenwich Mean Time has meant UT1 (see Section 2.24). Thus GMT has two
meanings that can differ by as much as 0!9. For these reasons GMT should not be
used for precise purposes.
2.53 Rotation of the Earth
The continued failure of successive lunar theories to represent the observed motion
of the Moon gradually led to the realization that the rotation of the Earth is not
uniform. In ephemerides calculated from gravitational theories, the tabular times
are the values of a uniform measure of time, quite independent of the variable
2 / TIME 77
rotation of the Earth. Consequently, the observed hour angle of a celestial body
(which increases with the Earth's rotation) differs from the position tabulated in
the ephemeris. An analysis of the discrepancies between observed and computed
positions reveals the difference between a measure of the Earth's rotation and the
measure of a uniform gravitational time. The discrepancies are most evident in the
mean motion of the Moon, due to the rapidity of its motion and the accuracy with
which the inequalities of motion can be observed. However, variations in the Earth's
rotation also produce discrepancies in the observed motions of the other bodies in
the solar system, in proportion to the magnitude of their respective mean motions.
The first variation to be recognized was a secular retardation of the rate of ro-
tation due to tidal interaction between the Earth and Moon. Adams (1853) showed
that the observed secular acceleration of the mean motion of the Moon could not
totally be produced by gravitational perturbations. However, the validity of his re-
sults was hotly contested. The issue was resolved when Ferrel (1864) and Delaunay
(1865) showed from dynamical principles that the tides exert a retarding action on
the rotation of the Earth, accompanied by a variation of the orbital velocity of the
Moon in accordance with the conservation of momentum.
Newcomb (1878) considered the possibility of irregular variations of the Earth's
rotation as an explanation of lunar residuals. In the end, however, he could not
find collaboration from planetary data (Newcomb, 1912). Some of the difficulties
faced by Newcomb, and prior astronomers, resulted from inadequacies in the lunar
theory. This situation was rectified when Brown (1919) introduced his new tables. A
number of studies offered increasing evidence of irregular variations in the Earth's
rotation. Most important were papers by de Sitter (1927) and Spencer Jones (1939),
which correlated irregularities in the Moon's motion with irregularities of the inner
planets.
Meanwhile, the accuracy of crystal-controlled clocks was becoming comparable
with that of the rotation of the Earth. By comparing the observed rates of the
clocks of national time services, Stoyko (1937) found a periodic seasonal variation
in the rate of rotation.
The Earth's rotation rate was therefore understood to have secular, irregular,
and periodic variations. Since mean solar time was determined directly from sidereal
observations, it could no longer be considered uniform.
2.54 Universal Time
Newcomb (1898) gave the following expression for the right ascension of the ficti-
tious mean Sun:
RG) = 18h38m45=836 + 86401 84?542r+0;0929r2,
(2.54-1)
78 EXPLANATORY SUPPLEMENT
where T is measured in Julian centuries from 1900 January 0, 0h Greenwich mean
noon. This expression differs from that of the mean longitude of the Sun by only
a slight, progressively increasing excess of 0?0203 T2. This difference is due to the
secular acceleration of the Sun and to the different rates of general precession on
the ecliptic and on the equator.
By inserting Equation 2.54-1 into the formula for mean solar time, Equa-
tion 2.52-1, and evaluating the result at 0h mean solar time on the Greenwich
meridian, we obtain the Greenwich mean sidereal time at 0h mean solar time:
GHAQo = 6h38m45*836 + 86401841542 T + 0?0929 T2, (2.54-2)
where T takes on successive values at a uniform interval of 1 / 36525. But because
the rotation of the Earth is not uniform, values calculated from this expression are
not consistent with observed hour angles of stars.
From the beginning of the century through 1959, these formulas were used
in the British Nautical Almanac and in the American Ephemeris to define mean
solar time and to relate solar time and sidereal time. As given by Newcomb, the
independent variable in Equation 2.54-1 is consistent with that of his solar tables.
As evidence accumulated that the Earth did not rotate uniformly, Newcomb's T
was interpreted as the independent variable of the Earth's orbital revolution rather
than its axial rotation. Since Newcomb's tables were the basis of the fundamental
ephemerides of the inner planets, the independent variable of his tables was called
ephemeris time (ET). It became associated with the concept of a uniform measure
of time, based on the laws of orbital dynamics.
To clarify notation, 7"e (expressed in Julian centuries of 36525 ephemeris days
from 1900 January 0, 12h ET) can be used in Equation 2.54-1 to signify ephemeris
time. Then #q(7e) = Re defines a uniformly moving fiducial point that is indepen-
dent of the Earth's rotation.
To make Equation 2.54-2 consistent with observed hour angles of the stars,
Universal Time (UT) was instituted as a measure of the Earth's rotation. Then
Equation 2.54-1 becomes a definition of UT, with 7\i (measured in Julian centuries
of 36525 days of Universal Time from 1900 January 0, 12h UT) serving as the
independent variable: Rq(Tv) -Rv. The difference between RE and Ry is given by
/?E - J?u = 0. 002738A7\ (2.54-3)
where AT is the difference between ephemeris time and Universal Time (AT —
ET - UT).
Prior to 1960, the designation "Right Ascension of Mean Sun+ 12"" was some-
times applied to tabulations of Greenwich sidereal time at 0h UT. This inexact
2 / time 79
terminology was dropped when the distinction between ephemeris time and Uni-
versal Time was introduced in national ephemerides in 1960. Equation 2.54-2, with
Tu as the independent variable, was used during the period 1960-1983 to calculate
tabular values of Greenwich mean sidereal time at 0h UT. Beginning with ephemer-
ides for 1984, Equation 2.24-1 has served as the defining relation between Universal
Time and sidereal time {see Section 2.24).
2.55 Ephemeris Time
From evidence compiled by de Sitter (1927) of fluctuations in the Earth's rotation,
Danjon (1929) recognized that mean solar time, which was observationally tied to
the rotation of the Earth, did not satisfy the need for a uniform timescale. Be sug-
gested using the timescale of the Newtonian laws of planetary motion. His call for
reform, appearing in a nontechnical journal, was perhaps unnoticed by specialists;
certainly it was unheeded. Two decades later, Clemence (1948) published a more
specific proposal, apparently without being aware of Danjon's paper (Clemence,
1971). From the results of Spencer Jones (1939), Clemence derived the factor AT
that would reduce Universal Time to the measure defined by Newcomb's tables. In
addition, Clemence determined the correction to Brown's lunar theory that would
make the independent variable of the lunar theory consistent with that of New-
comb's Sun. Thus a comparison of lunar observations with Brown's (corrected)
theory could be used to determine AT.
In 1948 a proposal to establish a more uniform fundamental standard of time
was referred by the Comite International des Poids et Mesures to the Interna-
tional Astronomical Union. This proposal was considered at the Conference on the
Fundamental Constants of Astronomy held at Paris in 1950. At this conference,
Clemence proposed using the measure of time defined by Newcomb's Tables of the
Sun. The Conference adopted a resolution recommending that this measure of time
be adopted, be expressed in units of the sidereal year at 1900.0, and be designated
by the name ephemeris time, as suggested by Brouwer. This recommendation was
adopted in 1952 by the International Astronomical Union at its General Assembly
in Rome.
Upon further consideration, the tropical year, which was considered more fun-
damental than the sidereal year, was chosen as the unit of time. Accordingly, the
Comite International des Poids et Mesures, at its session in September 1954 in
Paris, proposed to the 10th General Conference on Weights and Measures that the
fundamental unit of time be the second, redefined as 1/31556925.975 of the length
of the tropical year for 1900.0. The Conference authorized the Comite to adopt
this unit after formal action on the definition had been taken by the International
Astronomical Union.
The IAU, at its General Assembly in Dublin in September 1955, approved the
definition proposed by the Comite. However, the tropical year was understood to
80 EXPLANATORY SUPPLEMENT
be the mean tropical year defined by Newcomb's expression for the geometric mean
longitude of the Sun; the value of the second required for exact agreement with
Newcomb's tables is 1/31556925.97474 of the tropical year. The primary unit of
ephemeris time was the tropical year at the fundamental epoch of 1900 January
0, 12h ET. The tropical year was defined as the interval during which the Sun's
mean longitude, referred to the mean equinox of the date, increased by 360°. The
adopted measure of this unit was determined by the coefficient of T, measured in
centuries of 36525 ephemeris days, in Newcomb's expression for the geometric mean
longitude of the Sun (referred to the mean equinox of date):
L = 279°41 ' 48 "04 + 1 29602768'.' 13T+ 1 "089 T2. (2.55- 1 )
The tropical year at 1900 January 0, 12h ET accordingly contained
360 x 60 x 60 ^ ^ Q = 1556925_ 9?47 ephemeris seconds. (2.55-2)
129602768.13
Consequently, the Comite at its session in Paris in October 1956, under the author-
ity given by the 10th General Conference, adopted in place of the value formerly
recommended a slightly more precise value of the tropical year at 1900 January 0,
12h ET: "La seconde est la fraction 1/31556925.9747 de l'annee tropique pour 1900
janvier 0 a 12 heures de temps des ephemerides," (Proces Verbaux des Seances,
1957).
At this session, a Comite Consultatif pour la Definition de la Seconde was estab-
lished to coordinate the work of physicists on atomic standards and of astronomers
on the astronomical standard of ephemeris time. In 1960, the 11th Conference Gen-
erale des Poids et Mesures formally adopted the ephemeris second as a fraction of
the tropical year.
At its 1958 General Assembly in Moscow, the International Astronomical Union
adopted the following definition for the epoch of ephemeris time: "Ephemeris time
is reckoned from the instant, near the beginning of the calendar year A.D. 1900,
when the geometric mean longitude of the Sun was 279°4r48'.'04, at which instant
the measure of ephemeris time was 1900 January 0d12" precisely."
With its basic unit and initial epoch defined in this way, ephemeris time was
equivalent to the system of time of Newcomb's solar tables. Newcomb originally
identified this timescale with mean solar time, which he considered to be uniform.
When mean solar time, as determined from observations, was shown to be nonuni-
form, Newcomb's timescale was identified directly with ephemeris time. The first
two terms of the Sun's geometric mean longitude were defined as absolute con-
stants, and the corresponding values for the Moon and other planets were subject
to possible revision to bring them into accord with observation. The symbol T in
these expressions was therefore intended to represent ephemeris time, not only in
2 / TIME
81
the theory of the Earth's motion around the Sun, but also in the heliocentric the-
ories of the other planets. Ephemerides derived from Newcomb's tables of the Sun
and planets could therefore be regarded as having ephemeris time as the indepen-
dent time argument. The mean longitude of any other planet, or even of the Moon,
could have been used to define the origin and rate of a uniform time system, and
ephemerides of the Sun, Moon, and planets could have been constructed with that
time system as the independent argument.
2.551 The Determination of Ephemeris Time In principle, the ephemeris time at
any moment can be determined by comparing the observed positions of the Sun,
Moon, and planets with then corresponding ephemerides. The tabular time for
which the observations agreed with the ephemerides would be the ephemeris time.
In practice, Universal Time, which can be determined very accurately and with
little delay from observations of the diurnal motions of the stars, was used as an
intermediary measure of time. The difference between the two measures of time,
AT - ET — UT, was then obtained by comparing observations with ephemerides.
Observations of the Moon, whose geocentric motion is much greater than those
of other bodies, provided the most accurate determination of AT. However, direct
comparison with the lunar ephemeris calculated from Brown's lunar tables did not
give AT immediately, since Brown's theory was not strictly gravitational and his
tables were not in complete accord with Newcomb's solar tables. This was rectified
by eliminating the empirical term from the mean longitude of the Moon and by
applying the correction to the tabular mean longitude:
AL = 8'.'72- 26.74r- 11.22T2. (2.551-1)
Additional corrections were needed for some of the periodic terms in longitude,
latitude, and parallax. Beginning in 1960, the lunar ephemeris was calculated from
this amended theory — directly from the theoretical expressions for the longitude,
latitude and parallax — instead of from Brown's tables. The improved ephemeris was
made available for 1952-1959 in the Improved Lunar Ephemeris (Nautical Almanac
Office, 1954).
A distinction had to be drawn between UT + AT and ET, when AT was deter-
mined using the previous equations and observations of the Moon. UT+AF differed
from ET in two main respects:
(1) by a quadratic expression in T of the form a + bT + cT2, the coefficients of
which had been observationally determined to be zero, but which almost
certainly differed from zero by significant amounts (it should be noted that
the term cT2 is more of a fundamental physical character than a + bT)\
(2) by any deficiencies that might have been present in Brown's lunar theory,
including revision of any constants involved; in particular, Brown used 1/294
for the flattening of the Earth.
Q2 EXPLANATORY SUPPLEMENT
In the period from 1960 to 1984, improvements were introduced in the lunar
theory, and the different versions of the lunar theory were designated by j = 0,
j = 1 and j = 2. The ephemeris times determined from these lunar theories were
correspondingly labeled ETO, ET1 and ET2.
Thus the value of A T could differ systematically depending upon which method,
or body, was used for its determination. This will continue to be true when AT is
extrapolated back in time prior to the availability of atomic time.
2.552 Difficulties in the Definition of Ephemeris Time In hindsight, the definition
of ephemeris time (ET) was subject to a number of difficulties, both in concept and
in implementation. It was based on Newcomb's theory of the Sun and an associated
set of astronomical constants. Both the theory and the constants were replaced in
1984. At a deeper level, however, ET suffered from its prerelativistic origins. Since
its definition ignored general relativity, ET cannot be categorized as a geocentric
or bary centric timescale, nor as a proper time or coordinate time.
Both the fundamental unit and the epoch of ET contained technical problems.
The tropical year was chosen as the unit of time under the assumption that it was
independent of astronomical constants, particularly the constant of precession. In
reality, determinations of both the tropical year and the sidereal year depend on
the adopted system of astronomical constants. Unfortunately, the possibility of a
time-dependent variation in the equinox, commonly called equinox motion, was not
recognized. As a result, ephemeris time was subject to corrections introduced in
the system of constants in 1984.
The fundamental epoch, designated 1900 January 0, 12h ET, was the instant
at which the geometric mean longitude of the Sun, referred to the mean equinox
of date, was 279°4r48!'04. Although this instant was definitive, its determination
depended on a comparison of observations with an apparent ephemeris of the Sun.
The observations were themselves definitive, but the apparent ephemeris as deduced
from the geometric mean longitude depended on the value adopted for the constant
of aberration. All relevant observations and determinations had been made using
20'.'47 for the constant of aberration. A change in this value would lead to a change
in the determination of the instant of the fundamental epoch and, thus, to a corre-
sponding change in the measures of ephemeris time assigned to all other instants.
This particular difficulty could have been avoided by specifying the epoch as the
instant when the geometric mean longitude of the Sun, reduced by the constant of
aberration and referred to the mean equinox of date, was 279°4r27'.'57; but there
were objections to the implied use of an "apparent mean longitude."
Ephemeris time was defined by the motion of a body that is difficult to observe.
Determination of ET was better accomplished by observations of rapidly moving
bodies — the Moon, Mercury, and Venus. However, there was no specified relation-
ship between ET as defined and ET as determined from observations of the Moon,
Mercury, or Venus. In practice, the determination of ET from the motion of the
2 / TIME 83
Moon was based on accepted values of various constants. Changes in the constants
resulted in different measures of ET, as described in Section 2.551.
2.553 AT Formulas from Historical Observations The variations in the rotation
of the Earth can be determined from observations, though early observations are
much less accurate than modern observations. The rate of rotation does not decrease
uniformly, as it would if tidal friction were the only mechanism affecting the Earth's
rotation.
The formulas for AT must be divided into time periods. The equation for the
time period from 1650 to the present is given by McCarthy and Babcock (1986):
A7XTDT-UT1) - 5!156(± 0.404)+ 13. 3066(± 0. 3264)(r-0. 19(± 0. 01))2, (2.553-1)
where ( is given in centuries since 1800.0. The equivalent expression for the length-
ening of the day is:
LOD = 7!286(± 0. 170) x I0_6(year - 1819. 25(± 1.04)). (2.553-2)
Tabular values of AT are given by McCarthy and Babcock (1986) for 1957-1984
and by Stepenson and Morrison (1984) for 1630-1980. Current values are given in
the The Astronomical Almanac.
The following expressions for historic periods are from Stephenson and Morrison
(1984). A parabolic representation of the data from A.D. 948 to A.D. 1600 is given
by
AT = 25*5^, (2.553-3)
where t is time in centuries from A.D. 1800. This result is equivalent to a rate of
lengthening of the day of 1.4ms/century.
For the period from 390 B.C. to A.D. 948 the parabolic representation is
AT = 1850s - 435r + 44. 3t2, (2.553-4)
where r is measured in centuries from A.D. 948. With the origin at A.D. 1800, the
corresponding expression is
AT= 1360s + 3201 + 44.3^. (2.553-5)
The equivalent rate of lengthening of the day is 2.4ms/century (Stephenson and
Morrison, 1984).
84 EXPLANATORY SUPPLEMENT
The formulas above are based on the value of the lunar tidal acceleration n,
where n = -26'.'00 century-2. If another value of n is used, the values of TDT - UT1
should be corrected by the addition of
A7TTDT - UT1) - -1!821<ZW2)(T- 19.55), (2.553-6)
where T is measured in Julian centuries from 1900.0 (McCarthy and Babcock,
1986).
2.56 History of Atomic Time
Until 1960, the unit of time, the second, was defined as a specified fraction (1/86400)
of the mean solar day. With the adoption of ephemeris time, the second was defined
to be a specified fraction of the year (see Section 2.55). This definition was more
precise than the definition based on the mean solar day, but it was difficult to
measure from observational data and to implement with operational clocks.
The idea of using atomic resonances in an atomic clock is due to Rabi, who
suggested it in an American Physical Society lecture in January 1945. The first
atomic clock in actual metrological use was built at the National Physical Labo-
ratory (NPL) by Essen and Parry (1957). Details of the early developments are
described by Ramsey (1972), Beehler (1967), and Foreman (1985).
Since 1967, the Systeme International (SI) second (see Section 2.21) has been
the standard unit of time in all timescales. The length of the SI second was defined
by Markowitz et al. (1958) in terms of the observationally determined value of the
second of ephemeris time. Thus, the SI second closely matches an ephemeris second
and provides continuity with ephemeris time.
Beginning in July 1955, atomic timescales were available from organizations in
several countries. These were based on the cesium-beam frequency standards at the
individual organizations. In 1971, the experimental atomic timescale established
by the Bureau International de l'Heure (BIH) was adopted as the worldwide time
reference system, under the name Internationa] Atomic Time (TAI). As this decision
implied no discontinuity of the BIH timescale, it is convenient to designate it as
TAI since its beginning in July 1955. The origin of TAI was arbitrarily chosen so
that the TAI and UT1 readings at 1958 January 1, 0h were the same.
Since 1971, TAI computation has undergone two major conceptual changes.
In 1973, as a consequence of the 1972 meeting of the Consultative Committee
for the Definition of the Second (CCDS), a new TAI algorithm was implemented,
whereby each clock participates with a weight that is a function of its past and
present frequency. The mean frequency of each clock over a two-month interval
is computed with respect to TAI. The weight of a clock is proportional to the
reciprocal of the variance of six mean frequencies. This takes into account the
changes of frequency or the short-term instability of the clock. The other change in
2 / TIME 85
TAI took place in 1977. In 1976, three laboratories maintaining cesium standards--
the National Bureau of Standards, National Research Council, and the Physicalisch-
Technische Bundesanstalt — agreed that the frequency of TAI was too high by 10-12.
That same year, IAU recommended that the TAI frequency be corrected by exactly
— 10 x 10-13 on 1977 January 1. This adjustment was the first direct input of the
laboratory standards on TAI.
In April 1977, the Consultative Committee for the Definition of the Second
recommended that frequency steering by frequent, small adjustments (of the same
order as the variations expected from random noise) was better than noticeable
corrections at less frequent intervals. This practice was applied immediately to
TAI.
In 1985 the IAU adopted a resolution agreeing to transfer the responsibility for
TAI from the BIH to the Bureau International des Poids et Measures, under the
responsibility of the International Committee on Weights and Measures (CIPM)
and the General Conference of Weights and Measures.
2.57 History of Coordinated Universal Time
Beginning in 1962, an increasing number of broadcast time services cooperated to
provide a consistent time standard. Most broadcast signals were synchronized to
the redefined UTC in 1972. Prior to 1972, broadcast time signals were kept within
0.1 second of UT2 by introducing step adjustments, normally of 0.1 second, and
occasionally by changing the duration of the second. The scale UT2 resulted from
applying the following formula for the seasonal variation in the rate of the Earth's
rotation to UT1:
UT2 = UT1 + 0!022 sin 2ttt - 0!012 cos 2?rr - 0?006 sin 4ttt + 0!007 cos47rr, (2.57-1 )
where r represents the fraction of the Besselian year of the observation.
The seasonal variation in the rotation of the Earth was first detected by means
of excellent crystal clocks at the Physicalisch-Technische Bundesanstalt (Scheibe
and Adelsberger, 1936). These variations were confirmed by Stoyko (1936, 1937),
Finch (1950), and Smith and Tucker (1953). Lunar tidal variations were found to
be in reasonable agreement with the theory. More recent reports are available by
O'Hara (1975), Guinot (1970), and Markowitz (1976). Previously, the research on
long-term variations in the rotation of the Earth was correlated with the motion
of the Moon, which was the only clock available with sufficient precision (Spencer
Jones, 1956; Munk and McDonald, 1960). Now UTC is based on UT1 as specified
in Section 2.24.
86
2.58 History of Transmitted Time Signals
EXPLANATORY SUPPLEMENT
Prior to 1956, the transmitted time signals were stepped or steered to maintain
a constant relationship to the rotation of the Earth. From 1956 to 1971, various
approaches were used by different organizations to maintain the transmitted signals
close to the value of time determined from the Earth's rotation. After 1971, all the
transmissions were tied to UTC as determined by the BIH. For the period 1956-
1971 it is possible to obtain a correction to the transmitted time signal from the
available documentation of the history of the transmission during that time period.
The Bureau International de PHeure, located at the Paris Observatory, was
founded to coordinate the practices followed by the national time services in ob-
servations and calculations for the determination of time, and to establish precise
international standards; it came under the auspices of the International Astronom-
ical Union in 1920.
In 1988, the BIH activities on astronomical time were taken over by a new
service with a wider scope — the International Earth Rotation Service (IER.S), the
Central Bureau of which is located at the Paris Observatory. The responsibility of
atomic timescales has been transferred from the BIH to the Bureau International
des Poids et Mesures (BIPM) in Sevres, France.
Table 2.58.1 provides the documentation of the steps and frequency changes
utilized by radio station WWV. In countries where the civil time system was not
the same as WWV, a different table of corrections appropriate to the civil time
system actually used should be introduced, if available.
Table 2.58.1
Time and Frequency Steps by WWV and MC (USNO)
Dale
MJD UT
Step*
Notes
1956
1957
Jan. 4
35476
1900
60 advance
Mar. 7
35539
1900
20 advance
Mar. 28
35560
1900
20
July 25
35679
1900
20
Aug. 22
35707
1900
20
Sept. 19
35735
1900
20
Oct. 31
35777
1900
20
Nov. 14
35791
1900
20
Jan. 23
35861
1900
20
Mar. 13
35910
2000
20
May 1
35959
1900
20
June 5
35994
2000
20
June 19
36008
1900
20
July 3
36022
1900
20
July 17
36036
1900
20
Aug. 14
36064
1900
20
Oct. 16
36127
1900
20
Nov. 6
36148
1900
20
Dec. 1 1
36183
1900
20
On 1 957 March 1 3, ihe time step made at 1 900 UT
was obtained from log books, but the step made at
2000 UT was from TS Bulletins.
Table 2.58.1, continued
Time and Frequency Steps by WWV and MC (USNO)
Date
MJD
UT
Step*
Notes
1958
Jan. 15
36218
1900
20
WWV controlled at an offset of about -100 x
Feb. 5
36239
1900
20
10~ 10 during 1958.
Feb. 1 9
36253
1900
20
Apr. 9
36302
1900
20
June 11
36365
1900
20
July 2
36386
1900
20
July 16
36400
1900
20
Oct. 22
36498
1900
20
Nov. 26
36533
1900
20
Dec. 24
36561
1900
20
1959
Jan. 28
36596
1900
20
WWV controlled at an offset of -100 x 10-10
Feb. 25
36624
1900
20
during 1 959.
Aug. 5
36663
1900
20
Aug. 26
36806
1900
20
Sept. 30
36841
1900
20
Nov. 4
36876
1900
20
Nov. 18
36890
1900
20
Dec. 16
36918
1900
20
USNO began controlling NBA at an offset of
-170x 10~10 during 1960.
1960
No time steps
UTC began 1960 January 1.t The initial partici-
pating observatories and laboratories were USNO,
RGO, NBS, NRL and NPL. The original offset was
-150 x 10-10. No international UTC offset was in
effect before 1960.
1961
Jan. 1
37300
0000
5
UTC otfset during 1961 was -150 x 10~10.
1962
Aug. 1
37512
0000
50 advance
UTC offset during 1962 was -130 x 10-10.
No time steps
1963
Nov. 1
38334
0000
100
UTC offset during 1963 was -130 x 10-10.
1964
Apr. 1
38486
oooo
100
UTC offset during 1964 was -150 x 10~10.
Sept. 1
38639
0000
100
Oct. 1
38669
0000
1
MC(USNO) advanced J. 6 milliseconds 1 0 October,
1964. WWV retarded 1.0 milliseconds 10 October,
1964.
1965
Jan. 1
0000
100
UTC offset during 1965 was -150 x 10~10.
Mar 1
38820
0000
100
July 1
38942
0000
100
Sept. 1
39004
0000
100
1966
No time steps
UTC offset during 1966 was -300 x 10~10.
1967
Sept. 20
39753
0000
200 p. advance
WWV only; MC(USNO) was not advanced 1967
1968 Feb. 1
1969
1970
1971
39887 0000 100 advance
No time steps
September 20. UTC offset during 1967 was -300 x
io-10.
UTC offset during 1 968 was -300 x 10
-10
UTC offset during 1969, 1970, and 1971 was
-300 x 10"10.
* All steps are retardations in milliseconds unless otherwise noted.
+ See Trans. IAU Reports, XI A, 362-364.
Note: Prior to 1 956 January 1 , WWV did not make time steps. Instead WWV steered its frequency to follow
closely the Earth's rotation. Steps were therefore unnecessary. (The system followed by WWV was the N2
system, used by the U.S. Naval Observatory from 1953 April 1 until 1955 December 31).
88
EXPLANATORY SUPPLEMENT
2.6 REFERENCES
Adams, J.C. (1853). "On the Secular Variation of the Moon's Mean Motion" Phil. Trans.
CXLIII, 397-406.
Allan, D.W. (1985). "Characterization Optimum Estimation and Time Prediction of
Clocks" Proceedings of the Seventeenth Annual Precise Time and Time Interval (PTTI)
Applications and Planning Meeting, pp. 45-66.
Allan, D.W. (1983). "Clock Characterization Tutorial" Proceedings of Fifteenth Annual
Precise Time and Time Interval (PTTI) Meeting, Washington, pp. 459-476.
Allan, D.W., Blair, B.E., Davis, D.D., and Machlan, H.E. (April 1972). "Precision and
Accuracy of Remote Synchronization via Network Television Broadcasts, LORAN-C,
and Portable Clocks" Metrologia 8, 64-72.
Anderie, R.J. (1976). "Polar Motion Determined by Doppler Satellite Observations" Re-
port to Commission 19. {Geodes Bull. Grenoble).
Andrewes, W. and Atwood, S. (1983). The Time Museum: An Introduction (The Time
Museum, Rockford, IL).
Anderson, O.D. (1975). Time Series Analysis and Forecasting; The Box-Jenkins Approach
(Butterworth, London).
Aoki, S., Guinot, B-, Kaplan, G.H., Kinoshita, H., McCarthy, D.D., and Seidelmann, P.K.
(1982). "The New Definition of Universal Time" Astron. Astrophys. 105, 359-361.
Ashby, N. (1975), "An Earth-Based Coordinate Clock Network" NBS Tech. Note 659 (US
Govt. Printing Office, Washington, DC),
Ashby, N. and Allan, D. (1979). "Practical Applications of Relativity for a Global Coor-
dinate Timescale" Radio Sci. 14, 649-669.
Audoin, C. (1984). "Prospects for Atomic Frequency Standards" Proceedings of the Six-
teenth Annual Precise Time and Time Interval (PTTI) Meeting, pp. 1-48.
Bagley, A.S. and Cutler, L.S. (1964). "A New Performance of the Flying Clock Experi-
ment" Hewlett-Packard Journal 15, 11.
Barnes, J.A. (1976). "Models for the Interpretation of Frequency Stability Measurements"
NBS Technical Note 683 (U.S. Govt. Printing Office, Washington, DC).
Barnes, J.A., and Allan, D.W. (1964). "Effects of Long-Term Stability on the Definition
and Measurement of Short-Term Stability" Proc. IEEE-NASA Symposium Short-
Term Frequency Stability NASA Spec. Publ No. 80, 119-123.
Becker, G. (1974). Time Scales Relativity: Report of Physikalisch-Technische Bunde-
sanstalt, and acts of the second Cagliari Meeting on Time Determination, Dissemi-
nation and Synchronization, 1974. Editor: Astron. Inst, of Cagliari, Italy.
Becker, G., Fischer, B., Kramer, G., and Muller, E.K. (1967). "Die Definition der Sekunde
und die aligemeine Relativitatstheorie" (The Definition of the Second and the Theory
of General Relativity). PTB-Mitt. 77, 111.
Beehler, R.E. (1967). "A Historical Review of Atomic Frequency Standards" Proc. IEEE
55(6), 792-805. See also Blair (1974), 85-109.
Beehler, R.E., Mockler, R.C., and Richardson, J.M. (1965). "Cesium Beam Atomic Fre-
quency Standards" Metrologia 1, 114-131.
Benavente, J., Besson, J., and Parcelier, P. (1979). "Clock Comparison by Laser in the
Nanosecond Range" Radio Sci. 14, 4, 701-706.
Besson, J, (1970). "Comparison of National Time Standards by Simple Overflight" IEEE
Trans. Instr. and Meas., IM-19, 4, 227-232.
BIH (1973). Annual Report of the Bureau International de I'Heure.
2 I TIME
89
BIH Circular D. (Bureau international de I'Heure, Paris, France),
Blair, B.E., ed. (1974). "Time and Frequency" (Red Book) Nat Bur. Stand (US) Monogr.
No, 140.
Bodily, L.N. (1966). "Correlating Time from Europe to Asia and with Flying Clocks"
Hewlett-Packard Journal 16, No. 8.
Box, G.E.P., and Jenkins, G.M. (1970). Time Series Analysis Forecasting and Control
(Holden-Day, San Francisco).
Brown, E.W. (1919). Tables of the Motion of the Moon (Yale University Press, New Haven,
CT).
Brunet, M., Freon, G., and Parcelier, P. (1977). "Comparison of Distant Atomic Clocks
with the SYMPHONIE Satellite" Proc. Symposium on Telecommunication Measure-
ment, organized by URSI (Lannion, France, 3-7 October 1977).
Buisson, J., McCaskill, T., Morgan, P., and Woodger, J. (1976). "Precise Worldwide Sta-
tion Synchronization via the NAVSTAR GPS, Navigation Technology Satellite (NTS-
1)" Proceedings of the Eighth Annual Precise Time and Time Interval Meeting.
Capitaine, N. and Guinot, B. (1985). "Anomalies of Some Tidal Waves in UT1" Geophys
J.R. Astr. Soc. 81, 563.; C.C.D.S. (1970). Comite Consultatif pour la Definition de la
Seconde, 5e Session (Bureau International des Poids et Mesures, Sevres); CCIR Docu-
ments, Study Group 7 (Standard Frequencies and Time Signals), Geneva, Switzerland.
Chi, A.R. and Byron, E. (1975). "A Two-way Time Transfer Experiment Using a Syn-
chronous Satellite" Proceedings of the Seventh Annual Precise Time and Time Interval
(PTTI) Applications and Planning Meeting, pp. 357-376.
Clark, T.A. (1972). "Precision Timing and Very Long Baseline Interferometry (VLBI)"
Proc. Fourth Annual NASA and Department of Defense Precise Time and Time In-
terval Planning Meeting GSFC Report X-814-73-72, pp. 74-89 (Goddard Space Flight
Center, Greenbelt, MD).
Clemence, G.M. (1948), "On the System of Astronomical Constants" Astron. J. 53, 169
179.
Cooper, R.S. and Chi, A.R. (1979). "A Review of Satellite Transfer Technology: Accom-
plishments and Future Applications" Radio Sci. 14, 4, 605-619.
Costain, CO, Daams, H., Boulanger, J.S., Hanson, D.W., and Klepczynski, W.J. (1978).
"Two-way Time Transfers between NRC/NBS and NRC/USNO via the Hermes (CTS)
Satellite" Proceedings of the Tenth Annual Precise Time and Time Interval Meeting.
Washington NASA Technical Memorandum 80250, pp. 585-600.
Costain, C.C., Boulanger, J.S., Daams, H., Hanson, D.W., Beehler, R.E., Clements, A. J.,
Davis, D.D., Klepczynoki, W.J., Veenstra, L., Kaiser, J., Guinot, B., Azoubib, P.,
Parcelier, P., Freon, G,, and Brunet, M. (1979). "Two-way Time Transfer via Geosta-
tionary Satellites NRC/NBS, NRC/USNO and NBS/USNO via Hermes and NRC/LPFT
(France) via Symphonie" Proceedings of the Eleventh Annual Precise Time and Time
Interval (PTTI) Conference NASA Conference Publication 2129, pp. 499-519.
Counselman, CO (1976), "Radio Astrometry" Annu. Rev. Astron. Astrophys. 14, 197-
214.
Danjon, A. (1929). "Le temps, sa definition practique, sa mesure" L'Astronomie xliii,
13-22.
Danjon, A. (1959). Astronomie Generate (Sennac, Paris).
Davis, D.T., Weiss, M., Clements, A., and Allan, D.W. (1981). "Unprecedented Syntoniza-
tion and Synchronization Accuracy via Simultaneous Viewing with GPS Receivers;
Construction Characteristics of an NBS/GPS Receiver" Proceedings of the Thirteenth
90
EXPLANATORY SUPPLEMENT
Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting
NASA Conference Publication 2220, pp. 527-544.
Delaunay, C.E. (1865). "Sur l'existence d'une cause nouvelle ayant une action sensible sur
la valeur de l'equation seculaire de la Lune" Compte Rendu des Seances de I'Academie
des Sciences 61, 1023-1032.
de Sitter, W. (1927). "On the Secular Accelerations and the Fluctuations of the Longitudes
of the Moon, the Sun, Mercury and Venus" Bull, of the Astron. Institues of the
Netherlands iv, 21-38.
Essen, L. and Parry, J.V.L, (1957). "The Cesium Resonator as a Standard of Frequency
and Time" Phil. Trans. R. Soc. London Series A 250, 45-69.
Farley et al. (1968). "Is the Special Theory Right or Wrong?" Nature 217, 17.
Ferrell, W. (1864). "Note on the Influence of the Tides in Causing an Apparent Secular
Acceleration of the Moon's Mean Motion" Proc. of the American Academy of Arts
and Sciences VI, 379-383.
Forman, P. (1985). "Atoinichson: the Atomic Clock from Concept to Commercial Product"
Proc. IEEE 73, 1181.
Frisch and Smith (1963). "Measurement of Relativistic Time Dilation Using Meason"
Amer. J. Phys. 31, 342.
Gerber, E.A. and Sykes, R.A. (1966). "State of Art: Quartz Crystal Units and Oscillators"
Proc. IEEE 54(2), 103-116. See also in Blair (1974a), where the authors give a brief
update of the more recent developments in a Part B.
Granveaud, D.M. and Guinot, B. (1972). IEEE Trans. Instr. and Meas. IM-21, 4, 396.
Granveaud, D.M. (1979). "Evolution of the International Atomic Time TAI Computa-
tion" Proceedings of the Eleventh Annual Precise Time and Time Interval (PTTI)
Applications and Planning Meeting NASA Conf. Pub. 2129, pp. 185-196.
Guinot, B. (1969). "Formation de l'echelle de temps coordonnee par le Bureau interna-
tional de l'heure" Actes du Colloque international de chronometrie A-20, Serie A,
Paris.
Guinot, B. (1974). "Review of Time Scales" Proceedings of the Sixth Annual Precise Time
and Time Interval (PTTI) Planning Meeting GSFC X-814-75-117 (Goddard Space
Flight Center, Greenbelt, MD).
Hafele, J. and Keating, R, (1972). "Around-the- World Atomic Clocks: Predicted Rela-
tivistic Time Gains" Science 177, 166-170.
Halley, E. (1695). "Some Account of the Ancient State of the City of Palmyra with Short
Remarks upon the Inscriptions Found There" Phil. Trans. 19, 160-175.
Hanson, D.W. and Hamilton, W.F. (1974). "Satellite Broadcasting of WWV Signals"
IEEE Trans, on Aerospace and Electronics System AES-10, 5, 562-573.
Hellings, R.W. (1986). "Relativistic Effects in Astronomical Timing Measurements" As-
tron. J. 91, 650.
Higa, W.H. (1972). "Time Synchronization via Lunar Radar" Proc. IEEE 60, 5, 552-557.
Hinteregger, H.F., Shapiro, I.I., Robertson, D.S., Knight, C.A., Ergas, R.A., Whitney,
A.R., Rogers, A.E.E., Moran, J.M., Clark, T.A., and Burke, B.F. (1972). "Precision
Geodesy via Radio Interferometry" Science 178, 396-398.
Hudson, G.E. (1964). "Spacetime Coordinate Systems" Actes du Congres international de
chronometrie (Lausanne, Switzerland).
Hudson, G.E., Allan, D.W., Barnes, J. A., Hall, R.G., Lavanceau, J.D., and Winkler,
G.M.R. (1969). "A Coordinate Frequency and Time System" Proc. 23rd Annual Fre-
quency Control Symposium USAE Com.
2 / TIME
91
Hurd, W.J, (1972). "An Analysis and Demonstration of Clock Synchronization by VLBI"
Proceedings of the Fourth Annual NASA and Department of Defense Precise Time
and Time Interval (PTTI) Planning Meeting, pp. 100-122.
IAU Fifteenth General Assembly (1973). "Report of Commission 31 Proceedings 1974
Trans. IAUXVB, 149.
IAU Sixteenth General Assembly (1976), Proceedings 1977 Trans L4C/XVIB, 58.
Ives, H.E. and Stillwell, G.R. (1938). "An Experimental Study of the Rate of a Moving
Atomic Clock" Opt. Soc. Am. 28, 215.
Jenkins, G.M. and Watts, D.G. (1968). Spectral Analysis and its Applications (Holden-Day,
San Francisco).
Jespersen, J.L., Kamas, G., Gatterer, L.E., and McDoran, P.F, (1968). "Satellite VHF
Transponder Time Synchronization" Proc. IEEE, 56, 7, 1202-1206.
Johnston, K.J. et al (1983). "Precise Time Transfer Using MK III VLBI Technology" Proc.
of Fifteenth Annual Precise Time and Time Interval (PTTI) Meeting, Washington,
443-455.
Johnston, K.J. (1974). "Radio Astrometry" Proceedings of Sixth Annual Precise Time and
Time Interval (PTTI) Conference, Washington, p. 373.
King-Hele, D.G. (1980). "The Gravity Field of the Earth" Phil. Trans., A 294, 317-328.
Klemperer, W,K. (1972). "Long-baseline Radio Interferometry with Independent Fre-
quency Standards" Proc. IEEE SO, 602-609.
Klepczynski, W.J. (1984). "Time Transfer Techniques: Historical Overview, Current Prac-
tices and Future Capabilities" Proceedings of the Sixteenth Annual Precise Time and
Time Interval (PTTI) Meeting, Washington, pp. 385-402.
Kleppner, D. et al. (1965). "Hydrogen Maser Principles and Techniques" Phys. Rev. 138,
4A, A972-A983.
Landes, D. (1983). Revolution in Time (Harvard University Press, Cambridge, MA).
Loeb, H.W. (1972). "Efficient Data Transformation in the Time Domain Spectrometry"
IEEE Trans. Instr. Meas. 21(2), 166-168.
Markowitz, W. (1977). "Timekeeping and Its Applications" Advances in Electronic and
Electron Physics 44, 33-87.
Markowitz, W., Lidback, C.A., Uyeda, H., and Muramatsu, K. (1966). "Clock Synchro-
nization via Relay II Satellite" IEEE Trans. Inst and Meas. IM-15, 4, 177-189.
Markowitz, W., Hall, R.G., Essen, L., and Perry, J.V.L. (1958). "Frequency of Cesium in
Terms of Ephemeris Time" Phys. Rev. Lett. 1, 105,
Martin, W. (1966). "Digital Communication and Tracking: Time Synchronization Exper-
iment" JPL- Space Programs Summary III, 61, 37-42.
Meeks, M.L., ed. (1976). "Astrophysics, Part C: Radio Observations" Methods of Experi-
mental Physics 12 (Academic Press, New York).
Misner, C, Thorne, K., and Wheeler, J. (1973). "Gravitation" (Freeman, San Francisco).
Moritz, H. and Mueller, I.I. (1987). Earth Rotation (Frederick Ungar, New York).
Moyer, T.D. (1981). "Transformation from Proper Time on Earth to Coordinate Time in
Solar System Barycentric Space-Time Frame of Reference" Celestial Mechanics 23, 1,
33-56 and 57-68.
Mueller, I.I. (1969). SpheHcal and Practical Astronomy (Frederick Ungar, New York).
Murray, J.A., Pritt, D.L., Blocker, L.W., Leavitt, W.E., Hooten, P.M., and Goring, W.D.
(1971). "Time Transfer by Defense Communications Satellite" Proc. of 25th Annual
Frequency Control Symposium, Atlantic City, New Jersey, pp. 186-193.
Nautical Almanac Office (1954). Improved Lunar Ephemeris 1952-1959 (U.S. Government
Printing Office, Washington, DC).
92
EXPLANATORY SUPPLEMENT
Newhall, XX, Williams, J.G., and Dickey, J.O, (1988). "Earth Rotation from Lunar Laser
Ranging" BIH Annual Report for 1987, D 47.
Oaks, O.J, (1985). "Time Transfer by Satellite" Proc. of Seventeenth Annual Precise Time
and Time Interval (PTT1) Applications and Planning Meeting, pp. 77-90.
Neugebauer, O. (1975). A History of Ancient Mathematical Astronomy ( Springer- Verlag,
New York).
Newcomb, S. (1878). "Researches on the Motion of the Moon" Washington Observations
for 1875.
Newcomb, S. (1895). "Tables of the Motion of the Earth on its Axis and Around the Sun"
Astronomical Papers of the American Ephemeris and Nautical Almanac IV, I.
Newcomb, S. (1912). "Researches on the Motion of the Moon, Part II" Astronomical
Papers of the American Ephemeris and Nautical Almanac IX, I.
Percival, D.B. (1973). Statistical properties of high performance cesium standards" Proc.
on the Fifth Annual Precise Time and Time Interval (PTTI) Conf., pp. 239-263.
Peters, H.E. and Washburn, P.J. (1984). "Atomic Hydrogen Maser Active Oscillator Cav-
ity and Bulb Design Optimization" Proc. of 16th Annual Precise Time and Time
Interval (PTTI) Meeting, Washington, pp. 313-338.
Potts, C.B. and Wieder, B. (1972). "Precise Time and Frequency Dissemination via the
LORAN-C System" Proc. IEEE 60, 530-539.
Pound, R. and Rebka, G. (1960). "Apparent Weight of Photons" Phys. Rev. Lett. 4, 337.
Pound, R. and Snider, J. (1964). "Effect of Gravity on Nuclear Resonances" Phys. Rev.
Lett. 13, 539.
Comite International des Poids et Mesures (1957). Proces Verbaux des Seances, deuxieme
serie 25, 75-79.
Ramsey, N.F. (1968). "The Atomic Hydrogen Maser" Am. Sci. 56(4), 420-438.
Ramsey, N.F. (1972). "History of Atomic and Molecular Standards of Frequency and
Time" IEEE Trans. Instram. and Meas. 21(2), 90-99.
Rapp, R.H. (1974). "Current Estimates of Mean Earth Ellipsoid Parameters" Geophys.
Res. Lett. 1(1), 35-48.
Recommendations and Reports of the CCIR 1978. "Standard Frequencies and Time Sig-
nals" VII, Report 363-4, 78-86.
Reder, F.H. and Winkler, G.M.R. (1960). "Preliminary Flight Tests of an Atomic Clock
in Preparation of Long-Range Clock Synchronization Experiments" Nature 186, 4725,
592-593.
Reder, F.H., and Winkler, G.M.R. (1960). "World Wide Clock Synchronization" IRE
Trans. Mil. Electron. 4(2/3), 366-376.
Reinhardt, V.S. (1974), "Relativistic Effects of the Rotation of the Earth on Remote Clock
Synchronization" Sixth Annual Precise Time and Time Interval (PTTI) Planning
Meeting, Washington, pp. 395-424.
Reinhardt, V.S. and Lavanceau, J. (1974). "A Comparison of the Cesium and Hydrogen
Hyperfine Frequencies by Means of Loran-C and Portable Clocks" Proceedings of the.
Twenty- Eighth Annual Symposium on Frequency Control, Ft. Monmouth, New Jersey,
379.
Reinhardt, V.S., Premo, D.A., Fitzmaurice, M.W., Wardrip, S.C., and Arvenka, P.O.
(1978). "Nanosecond Time Transfer via Shuttle Laser Ranging Experiment" Proceed-
ings of the Ninth Annual Precise Time and Time Interval (PTTI) Applications and
Planning Meeting, pp. 319-342.
2 / TIME
93
Ringer, D.E., and Gandy, J., and Jechart, E. (1975). "Spaceborne Rubidium Frequency
Standard for Navstar GPS" Proceedings of the Seventh Annual Precise Time and
Time Interval (PTTI) Conference, pp. 671-696.
Rogers, A.E.E. (1970). "Very Long Baseline Interferometry with Large Effective Band-
width for Phase Delay Measurements" Radio Sci, 5, 1239-1248.
Rueger, L.J. and Bates A. (1978). "Nova Satellite Experiment" Radio Sci. 14, 707-714.
Rutman, J. (1978). "Application of Space Techniques to Time and Frequency Dissemintion
and Synchronization" Proc. European Workshop on Space Oceanography, Navigation,
and Geodynarnics (SONG), Schloss-Elmau, Germany ESAPS-137. Paris: European
Space Agency SP-127, pp. 311-316.
Saburi, Y. (1976). "Observed Time Discontinuity of Clock Synchronization in Rotating
Frame of the Earth" J. Radio Res. Labs. 23, 112, 255-265.
Saburi, Y., Yamamoto, M,, and Harada, K. (1976). "High-precision Time Comparison
via Satellite and Observed Discrepancy of Synchronization" IEEE Thins. Instr. and
Meas. IM-25, 4, 473.
Silverberg, E.C. (1974). "Operation and performance of a Lunar Laser Ranging Station"
Appl. Opt 13, 565-574.
Spencer Jones, H. (1939). "The Rotation of the Earth and the Secular Accelerations of
the Sun, Moon and Planets" Month. Not. R. AS 99, 541.
Steele, J. McA., Markowitz, W., and Lidback, C.A. (1964). "Telstar Time Synchroniza-
tion" IEEE Trans. Instr. and Meas. IM-13, 4, 164-170.
Stoyko, N. (1937). "Sur la periodicite dans l'irregularite de la rotation de la Terre"
Compt.es- rendus 250, 79.
Stoyko, M.N. (1937). "Sur la periodicite dans l'irregularite de la rotation de la terra" C.R.
Acad. Sci 205, 79.
Taylor, R.J. (May 1974). "Satellite to Ground Timing Experiments" Proc. 28th Ann. Sym.
on Frequency Control, pp. 384-388.
Trans. IAU (1960). 10, 72, Moscow, 1958 (English translation, p. 500).
Turner, A.J, (1984). "Time Measuring Instruments: Water-clocks, Sand-glasses, Fire-
clocks" Catalogue of the Collection, I, Part 3 (The Time Museum, Rockford).
Vanier, J. and Audoin, C. (1989). "The Quantum Physics of Atomic Frequency Standards"
(Adams and Hilger, Bristol).
Veillet, C. and Pham Van, J. (1988). "Earth Rotation from Lunar Laser Ranging" BIH
Annual Report for 1987 D-45.
Vessot, R.F.C. (1974). "Lectures on Frequency Stability of Clocks in the Gravitational
Redshift Experiment" Experimental Gravitation B. Bertotti, ed. (School of Physics
Enrico Fermi, Academic Press, New York), pp. 111-162.
Vessot, R.F.C. and Levine, M.W. (1979). "A Test of the Equivalence Principle Using a
Space-Borne Clock" General Relativity and Gravitation 10, 3, 181-204.
Whitrow, G.J. (1988). Time in History (Oxford University Press, Oxford).
Winkler, G.M.R. (1977). "Timekeeping and Its Applications" Advances in Electronics and
Electron Physics 44, 33-97.
Winkler, G.M.R. (1987). "High Performance Clocks" First European Time and Frequency
Forum, 18-20 March 1987, Besancon, France, pp. 7-19.
Winkler, G.M.R. and Van Flandern, T.C. (1977). "Ephemeris Time, Relativity and the
Problem of Uniform Time in Astronomy" Astron. J. 82, 1, 84-92.
Woolard, E.W. and Clemence, G.M. (1966). Spherical Astronomy (Academic Press, New
York and London),
CHAPTER
Celestial Reference Systems
by C.Y. Hohenkerk, B.D. Yallop, C.A. Smith, and AT. Sinclair
3.1 CELESTIAL REFERENCE SYSTEMS
In order to specify positions of astronomical objects, such as stars, galaxies, quasars,
planets, and satellites, it is necessary to have a reference system. The currently avail-
able reference system is based on the FK5 star catalog, which provides a reference
frame, the IAU 1976 System of Astronomical Constants, the IAU 1980 theory of
nutation, and the procedures that provide the methods for implementing the ref-
erence system. Chapter 4 describes in detail the distinction between a reference
system and a reference frame.
An ideal celestial system would be an inertial system so that the equations
of motion could be written without any rotational terms. Unfortunately, the re-
alization of an ideal celestial reference system can only be approached, it cannot
be achieved. As this chapter was being prepared, the IAU appointed a working
group to study the many different aspects of reference systems and to formulate
improvements for the future. Therefore, although this chapter attempts to provide
the instructions and the methods for the current reference system, it also indicates
some of the difficulties with the present system and possible future changes.
The ideal reference system would provide an inertial frame that could be used at
all frequencies of the electro-optical spectrum, and at all magnitudes or intensities.
Unfortunately a number of problems interfere with realizing such a reference system.
The first problem is that no known point sources radiate at all frequencies of the
spectrum such that the positions determined at one frequency can be accurately
related to those determined at another frequency. Thus we have different sets of
reference somc.es in the optical, radio, and X-ray frequencies.
We have the problem that the dynamical range for all observing systems is
limited. Generally in the optical range only five magnitudes of intensity can be
observed astrometrically by one technique. Thus the method and the references
95
96 EXPLANATORY SUPPLEMENT
used for observing bright stars are different from those for observing faint stars. We
can define a reference frame based on the very distant extragalactic sources that
should be without proper motion and provide a pseudofixed reference frame, but
it is impossible to measure the positions of brighter stars or solar system objects
directly relative to those faint distant sources. Thus we have a problem realizing an
observable reference frame at all frequencies and magnitudes.
There is also a problem defining the ideal reference frame. In classical mechanics
an inertial reference frame is either at rest or in a state of uniform rectilinear
motion with respect to absolute space. However, the theories of general and special
relativity required some changes to this concept . In the general theory of relativity,
Einstein defined an inertial frame as a freely falling coordinate system in accordance
with the local gravitational field which is due to all material matter of the universe
(Einstein, 1956). The special theory of relativity defines an inertial frame in a
space-time continuum in the absence of gravitational fields, rather than in absolute
space (Moritz, 1967), and the transformation between inertial frames is given by
the Lorentz transformation.
There are finite regions with respect to a chosen space of reference where ma-
terial particles move freely without acceleration and in which the laws of special
relativity hold with remarkable accuracy, which can be regarded as quasi-inertial
frames. Thus the center of mass of the Earth-Moon system falling in an elliptic
orbit around the Sun in a relatively weak gravitational field is an example of such
a finite region. Therefore, we may assume quasi-inertial frames of reference, and
any violation of the principles when using classical mechanics can be taken into
account with small corrections applied to the observations and by an appropri-
ate coordinate-time reference. The effects of special relativity for a system moving
with the Earth around the Sun are of the order 10~8; likewise those with general
relativity are of the order 10-8.
It is necessary to distinguish between various types of quasi-inertial frames.
For example, the galactic reference frame, which is based on assumptions about the
proper motions of celestial bodies and their statistical properties, is a frame defined
kinetically. A reference frame can be constructed on extragalactic sources based on
the assumption that the galaxies have no rotational components to their motions.
Alternatively, we can construct a quasi-inertial frame based on a dynamical system.
This reference frame is based on the theory of motions of the bodies of the solar
system, and is constructed in such a way that there remains no rotational term in
the equations of motion.
In Newtonian mechanics, the various types should be equivalent, but this is not
true in general relativity. In general relativity a dynamical system of coordinates
is a local reference frame that is locally tangent to the general space-time mani-
fold. In contrast the kinetic frame is defined by assumptions about the apparent
motions of remote objects and is therefore a coordinate system that is subject to
relativistic effects such as geodesic precession. Even if this is suitably corrected, a
3 / CELESTIAL REFERENCE SYSTEMS 97
basic difference between the concepts remains, which is another reason for using
the terminology quasi-inertial.
Although we may desire a single ideal reference system, in practice we have to
recognize not only the conceptual differences in reference frames used in the avail-
able systems but also the differences in the realizations of these frames. Although
the FK5 can be viewed as a kinetic frame based on the stars, its origin is tied to
an equinox that is defined by the solar system. But, in practice, the FK5 does not
have a simple fiducial point. Its real origin is a meridian of zero right ascension.
The meridian does not exactly go through the equinox defined by a given solar sys-
tem ephemeris, nor is that meridian necessarily a plane. It can differ systematically
in a declination-dependent way. Therefore, the origin of the FK5 reference frame
is defined as a catalog equinox. This is contrasted with the equinox defined by a
solar-system ephemeris and called a dynamical equinox. Each solar-system ephem-
eris defines its own origin or equinox. In the case of the ephemerides DE200/LE200,
an effort was made to make the origin agree with the FK5 equinox, but it was rec-
ognized, and needs to be recognized, that there is some difference, which will be
determined more accurately with time.
Therefore, although we may have a conceptual ideal reference system, each
realization of a reference frame will be somewhat different, and there will be a need
to determine the transformations from one reference frame to another. The FK5
reference frame is very close to the DE200/LE200 reference frame, which in turn
is very close to the radio-source reference frames. The deviations are partly due
to the different accuracies with which the various reference frames are determined.
Similarly there can be time-dependent differences between the reference frames.
The FK4 reference frame was found to have a motion as a function of time. This
could be viewed as a systematic error in the proper motions of the FK4 system.
Although it is hoped that these differences will be small, it must be recognized
when doing precise positional astronomy that these differences do exist.
For precise reduction of observations it is necessary to include all the displace-
ments and motions involved in the reference systems. These include the differences
in the origin of the frame, barycentric, geocentric, or topocentric, together with the
appropriate time coordinate as discussed in Chapter 2. Also the time-dependent
variations in the celestial coordinates must be included; these are precession, nuta-
tion, parallax, aberration, and light deflection. The reduction to topocentric coor-
dinates includes the Earth-orientation parameters (UT1-UTC and polar motion),
diurnal parallax, diurnal aberration, and refraction. The basis for the expressions
necessary for these conversions is the subject of the rest of this chapter.
3.1 1 Fundamental Reference Systems
Stellar reference frames can be determined observationally. Because of the many
motions involved, it must be recognized that the observational determination of the
98 EXPLANATORY SUPPLEMENT
reference frame depends upon the dates of the observations. From the rotation of the
Earth, the plane of the equator and hence its pole can be determined. A fiducial
direction in the equatorial plane must be determined either from the definition
of a given object, collectively from a number of different objects, or from solar-
system observations defining the ecliptic. The observational determination of the
fundamental reference frame can then be converted to a conventional standard
reference frame by application of certain specified procedures. These procedures, in
combination with the reference frame, constitute the reference system and will be
specified in this chapter.
3.12 The Dynamical Reference System
The motions in the solar system provide a number of possible bases for reference
frames, the standard being the equator of the Earth and the equinox defined by the
intersection of the mean planes of the equator and the ecliptic. Alternatively, the
ecliptic and the equinox can define a reference frame. Some would argue that the use
of the invariable plane with an equinox provides a suitable reference frame, but this
frame is not well-determined. A reference frame, however defined, can be determined
from the process of calculating the ephemerides of the solar-system bodies based
on their equations of motion in the chosen frame, fitting these computed positions,
suitably reduced to the frame of observation, to observed positions, and hence
determining computed ephemerides that are consistent with the observational data.
Such a fitting process, with sufficiently accurate observation data, can also provide
the basis for determining the constants for the procedures necessary for reducing
the observational data. Some observing techniques for solar-system objects, such as
range measurements and spacecraft observations, are independent of any external
frame, such as the stellar frame, but these observations have a strong dependence
on the observing sites, i.e., the terrestrial coordinate system, the Earth-orientation
parameters, and an accurate basis for time. Other observations are made with
reference to stellar coordinate frames, for which the dynamical reference frame
must be determined with reference to the stellar reference frame for the consistent
use of these observations.
3.13 The Conventional Celestial Reference System
Conventional reference systems can be defined to provide a consistent basis for
comparison. Such conventional reference systems must include specification of the
reference frame at a given date, and all the necessary procedures and constants
required to transform this conventional frame from one date to any other date.
3 / CELESTIAL REFERENCE SYSTEMS
3.2 BASIS OF REDUCTION OF CELESTIAL COORDINATES
99
A necessary part of any reference system is the procedures required to relate a
reference frame for one time to one for another time. These procedures involve
precession, nutation, space motion, parallax, aberration, and light deflection. In
this section, these different effects will be discussed from a theoretical basis along
with an explanation of the determination of these effects. To complete the discus-
sion, formulas for polar motion and refraction, which are required in topocentric
reductions, are given at the end of this section.
3.21 Precession
Since the plane of the ecliptic and the plane of the equator are used as planes of
reference, and their intersection, the equinox, is used as the fiducial point, it is
necessary to include the fact that these planes are in motion. The motion of the
ecliptic is due to the gravitational action of the planets on the Earth's orbit and
makes a contribution to precession known as planetary precession. If the equator
were fixed, this motion would produce a precession of the equinox of about 12" per
century and a decrease in the obliquity of the ecliptic of about 47" per century.
The motion of the equator is due to the torque of the Sun, Moon, and planets on
the dynamical figure of the Earth. It can be separated into two parts, the lunisolar
precession, which is the smooth, long-period motion of the mean pole of the equator
about the pole of the ecliptic, with a period of about 26,000 years, and nutation,
which is the short-period motion of the true pole around the mean pole with an
amplitude of about 9" and a variety of periods of up to 18.6 years. The combination
of lunisolar and planetary precession is called general precession.
The notation used here follows Lieske et al. (1977), who give a useful table that
contains cross-references to the notation used by other authors.
A basic way of describing the motion of the ecliptic is to use angles irA and
IIA, where irA is the angle between the mean ecliptic at a fixed epoch ef = to and
the mean ecliptic of date eD = /, and TIA is the angular distance from the equinox
T0 at the fixed epoch to the ascending node where the ecliptic of date meets the
fixed ecliptic of epoch (Figure 3.21.1). Expressions for nA and UA are obtained
from Newcomb's theory of the Earth's orbital motion about the Sun, but with
modifications to allow for improved determinations of the planetary masses. The
form of the expressions from the theory are
sin tta sin TIA = st + s\ t2 + Sz r3,
sin nA cos II A — ct + Ci r + c^t1 ',
(3.21-1)
100
EXPLANATORY SUPPLEMENT
Figure 3.21.1
The ecliptic and equator at epoch and date
where for continuity tta > 0 for r > 0 and tta < 0 for t < 0. Numerical values for the
coefficients s, si, S2, c, c\, and q are given in Table 3.211.1 (page 104), expressed
conventionally in arc seconds.
The precessional effect due to the motion of the ecliptic is described by the arc
PT in Figure 3.21.1, and is referred to as the accumulated planetary precession.
Xa- Its rate of change at epoch to is
X = sec-sec er,! (3.21-2)
where e0 is the obliquity of the ecliptic at the epoch to-
The basic parameter for describing lunisolar precession is Newcomb's preces-
sional constant, which has the value Po at epoch to- It is not quite constant but has
a very slow rate of change P\ of approximately -0'.'00369 per Julian century. The
constant occurs in the dynamical equations of motion for the equator because of the
torque produced by the Sun and Moon; so far it has been impossible to calculate
a sufficiently accurate value of P0 from its theoretical dependency on geophysical
parameters, and so the value used comes from the observed rate of precession.
Lunisolar precession at epoch f0,
i> = Po cos e0 - Ps, (3.21-3)
is the rate of change in longitude $4 along the ecliptic due to the motion of the
equator at epoch to, where Pg is called the geodesic precession, which is a relativistic
nonperiodic Coriolis effect amounting to 1 '.'92 per Julian century (Lieske et ai,
1977).
3 / CELESTIAL REFERENCE SYSTEMS 1 01
The combined effects of planetary precession and lunisolar precession is called
general precession in longitude. Lieske et al. (1977) have adopted the definition
of the accumulated general precession Pa as denned by Newcomb, shown in Fig-
ure 3.21.1, of
Pa = TR - T0R. (3.21-4)
Newcomb actually adopted an expression for pA that computed the angle T0Q, in
Figure 3.21.1, where Q is the projection of the equinox of date on the ecliptic at
the fixed epoch t0. The difference is small, being 0'.'0005 / cy2; however, at epoch tQ
both procedures give the same rate
P = tp ~ X cos e0 = P0 cos e0 - PK-s cot en, (3.2 1-5)
and p may be resolved into general precession in right ascension, m, and general
precession in declination, n, where
m - ip cos en - x = (Po cos e0 - Pg) cos e0 - s cosec e0,
n = ip sin eo = (A) cos en _ Pg) sin en;
hence p = m cos eo + n sin e0. (3.21-6)
The constants of precession in the past have been derived from the analysis
of the proper motions of stars. In the solar neighborhood the stars are themselves
revolving about the center of the galaxy. The method was thoroughly reviewed
by Fricke (1977), who concluded that Newcomb's lunisolar precession should be
corrected by +1'.'10 per Julian century with an uncertainty of ± 0"15 per Julian
century.
There are several ways of formulating precession. For example, the vector di-
rections of the polar axis of the Earth and the polar axis of the ecliptic may be
expressed in terms of polynomials involving time arguments (Fabri, 1980; Mur-
ray, 1983), which are then used to calculate the effects of precession. The method
discussed below has been adopted by the IAU (Lieske et ai, 1977) for rigorous
calculations. It has developed from the time of Newcomb, and is well-suited for
the practical application of corrections for precession to coordinates and orbital
elements.
The accumulated precession angles £4, zA, and 0A, which are used to calculate
the effect of precession on equatorial coordinates, are referred to a base epoch
and have time arguments that describe precession from an arbitrary fixed epoch
sf to an epoch of date ep. The angles are shown on the surface of a sphere in
Figure 3.21.2, and polynomial expressions for them are given in Section 3.211.
102
EXPLANATORY SUPPLEMENT
Figure 3.21.2
The precession angles za, and 9a
In the figure the pole of the equator at Bf is point Pq and the pole of the
ecliptic is point Nq. At the epoch eq, Pq has moved to P and Nq to N. Imagine a
right-handed set of three-dimensional Cartesian coordinate axes with origin at the
center of the sphere, x-axis pointing to To, where the equator and ecliptic meet (the
equinox) at the fixed epoch ep . The y-axis is 90° away in an easterly direction along
the equator, and the z-axis points toward the pole Pq. Initially precession will move
Pq toward To, but the movement of Nq as well will cause Pq to move in a slightly
different direction.
A rotation of —(A about the z-axis makes PqTq pass through P. This great circle
meets the mean equator of epoch at right angles at point Q and the mean equator
of date at right angles at R. The jc-axis is now in the direction Q; the y-axis points
toward the node S, where the two equators cross.
A rotation +8 a equal to the angular separation of P from Pq about the new
y-axis brings the mean equator of epoch to the mean equator of date. The z-axis
points to the pole of date P, the y-axis to the node 5, and the *-axis now points
toward R in the plane of the equator of date.
Finally, a rotation of — zA equal to the angle TPR about the z-axis brings R to
T, so that the jc-axis points toward T, the equinox of date, and still lies in the plane
of the mean equator of date.
3 / CELESTIAL REFERENCE SYSTEMS
103
The precession matrix P, made up of these rotations (see Section 11.4), precesses
equatorial rectangular coordinates from an arbitrary fixed equinox and equator of
epoch £f to one of date eo and is given by
P[eF , ed] = Rs(-^) Rz<+^) R3(-C0.
(3.21-7)
Rewriting Equation 3.21-7 in terms of spherical coordinates gives
P =
cos za cos 6a cos Ca
- sin za sin Ca
sin zA cos 9A cos Ca
+ cos za sin Ca
sin 6 A cos Ca
— cos zA cos 6a sin Ca ~ cos zA sin 8 a
- smzA cosGt
sin zA cos 8 a sin Ca
+ cos za cos Ca
■ sin 9 A sin Ca
— sin za sin 9A
COS #4
(3.21-8)
Having calculated the precession angles for the matrix P, one can calculate the
inverse matrix P_l in various ways; for example,
P"1 =R3-1(-G)Rr1 (+^)R3_ (~za)
= rI(-Ca)R2(+0a)rJ(-^)
= R3(+G)R2(-0a)R3(+zjO
= P[£D,£fI
(3.21-9)
where use has been made of the property that the inverse of a rotation matrix is
its transpose (i.e., it is orthogonal).
3.211 Precession Angles and Rates Adopted by IAU (1976) The new basis for pre-
cession is taken directly from the discussion of Lieske et al. (1977) and modified
by the small amount discussed in Lieske (1979). The following values are adopted
constants at epoch J2000.0
p = 5029'.'0966 per Julian century
P\ = — 0"00369 per Julian century
Pg = 1 '.'92 per Julian century
e0 = 23°26,21'.,448 (3.211-1)
104
EXPLANATORY SUPPLEMENT
Table 3.211.1
Accumulated Precession Angles
sin ka sin nA = (4"f 976 - 0:75250 T + 0:'000431 T2) t + (0"1 9447 + CT000697 T) fi - 0'.'0001 79 fi
sin it A cos nA = (-46:'81 50 - 0:001 17 7 + 0:'005439 T2)t + (0:'05059 - 0"0037 127)12 + 0:'000344 fi
tta = (47"0029 - 0"06603 T + 0:'000598 t2) t + (-Q'.'03302 + 0:000598 T) fi + 0!'000060 fi
nA = 174° 52 34!'982 + 3289:4789 7 + 0'.'60622 T2 + (-869'.'8089 - 0:'50491 T) f + 0"03536 t2
i-A = (5038'.'7784 + 0!'49263 T - O'.'OOOI 24 T2) t + (-1 !'07259 - O"001 106 T) t2 - 0"001 147 fi
XA = (10','5526 - 1 '.'88623 T + 0:'000096 T2) t + (-2:38064 - 0"000833 T)fi - 0!'0Q1 125 fi
pA = (5029:0966 + 2l'22226 T - 0!'000042 T2) t + (1 "1 1 1 13 - 0!'000042 T) fi - 0:'000006 fi
(A = (2306:2181 + 1'.'39656 T - 0'.000139 T2) t + (0:'30188 - 0'.'000344 T) fi + 0!'017998 fi
zA = (2306'.'21 81 + 1 :'39656 T - OI'OOOI 39 T2) t + (1 :09468 + 0'.'000066 7) fi + 0!'01 8203 fi
6A = (2004:3109 - 0!'85330 T - 0'.'00021 7 T2) t + {-0!42665 - O'.'00O21 7 T) fi - 0^041833 fi
The equations in Table 3.211.1 give expressions for the accumulated precession
angles as functions of time, where the base epoch of the equations is e0 = J2000. 0
or JD(eo) = 2451545.0, and the time arguments are in units of a Julian century, i.e.,
T = (JDOf ) - JD(e0)) / 36525
t = (JD(£D) - JD(eF)) / 36525, (3.21 1-2)
where eF is a fixed epoch and eo is the epoch of date.
The expressions at epoch £f for the rates per Julian century of general preces-
sion, in longitude, (/>), right ascension (m) and declination («), and ir the rate of
rotation of the ecliptic, are
p = x(P4)U = 5029"0966 + 2'.'22226 T - 0"000042 T2
at
m = 4(C/» +^)|f=o = 4612"4362 + 2'.'79312 7" - O'.'000278r2
at
n = 4(&)l*=o = 2004'.'3I09 - 0'.'85330r - O'.'0OO217 T2
T = -t(tta)\,=o - 47!'0029 - 0'.'06603 T+ 0!'000598 T2.
at
3.212 Rigorous Reduction for Precession The most convenient method of rigor-
ously precessing mean equatorial rectangular coordinates is to use the precession
matrix P as follows:
r = Pr0
and r0 = P_1r,
(3.212-1)
3 / CELESTIAL REFERENCE SYSTEMS 1 05
which transforms the position vector To referred to the fixed epoch ejr = tQ, to the
position vector r referred to the epoch of date eD = f, and vice versa. The matrix
P (see Equation 3.21-8) is evaluated using the precession angles Ct> Zai and 0A (see
Table 3.211.1) and the appropriate time arguments.
Equation 3.212-1 can be rewritten in terms of right ascension and declination
(a, 6) as
sin(cv - zA) cos 6 = sin(ao + CO cos 6n
cos(q - za) cos 6 = cos(a0 + CO cos 6A cos <50 - sin 9A sin So (3.212-2)
sin 6 = cos(q0 + CO sin 6A cos 8Q + cos 6A sin <50,
and the inverse is
sin(a0 + CO cos *o = + sin(a - za) cos 6
cos(a0 + C0cos<50 = +cos(a - za)cos6a cos<5 + sin^ sin 6 (3.212-3)
sin 60 = - cos(a - zA) sin 6A cos 6 + cos 9A sin 6.
3.213 Approximate Reduction for Precession Approximate formulas for the re-
duction of coordinates and orbital elements referred to the mean equinox and equa-
tor or ecliptic of date (f) are
For reduction to J2000.0 For reduction from J2000.0
a0
— a
— M — N sin am tan 8m
a
= ao + M + N sin am tan Sm
60
= 6-
- N cos am
6
= 60+N cos am
A0
= A
- a + bcos(\ + c') tan .in
A
= A0 + a — b cos( A0 + c) tan 3
/3b
= /3
- fcsin(A + c')
0
= 0o + i>sin(A0 + c)
r?0
= Q
— a + b sin(Q + c') cot z'o
Q
= Qq + a - b sin(Q<) + c) cot i
h
= i -
- bcos(Q + c')
i
~ io+b cos(Qu + c)
— U)
— b sin(£2 + c') cosec to
= ujQ + b sin(£in + c) cosec ('.
The subscript zero refers to epoch J2000.0, and am and 8m refer to the mean epoch;
with sufficient accuracy
am = a - ^(M + N sin a tan 5)
5m = 6 — -N cos am
106
EXPLANATORY SUPPLEMENT
or Qm = do + -(M + N sin a0 tan 6a )
Sm = 60 + ^N cos am (3.213-1)
and the precession angles, obtained to sufficient accuracy, from expressions in Ta-
ble 3.211.1 for the fixed epoch of J2000.0, T- 0, are
tta sin nA * 4" 1 976 1 + 0'.' 19447 f2 - 0'.'000179 P
tva cos nA = -46?8 150? + 0'.'05059 r2 + 0:'000344 13
M = (Q + ZaW = 0, /) = 1?28 12323 1 + 0?0003879 12 + 0?0000101 /3
N = $A(T = 0, t) = 0?5567530 1 - 0?000 1 1 85 r2 - 0?00001 1 6 13
a=pA(T=Q,t)= 1 ? 396971 f + 0?0003086 r
b = <rcA(T=Q,t) = 0?0 13056/ - 0? 0000092 f2
c= 180° -nm + -a = 5?12362 + 0?241614r + 0f0001122r2
c' = 180° -n„, - ^a = 5?12362- 1 ° 155358 1- 0?0001964f2 (3.213-2)
where 17m - IIA(T - j t, t - 0) is evaluated at the mid-epoch.
Formulas, valid over short periods, for the reduction from the mean equinox
and equator or ecliptic of epoch t\ (e.g., the middle of the year) to date (t = t\ +t),
where r is a fraction of a year and |r| < 1, are
a = «i + r(m + n sin at\ tan 6[) 6 = 6\ + rncos Q|
A = A, + r{p - 7tcos(A, +6°)tan^) P = P\ +T7rsin(A, +6°)
(3.213-3)
{2 = Q\ + r(p - 7T sin(£2i +6°)cotr) i = i\ +T7rcos(f2] +6°)
u - W] + r7r si in L 2| + 6°) cosec j
where p, m, n, and n are the annual rates, evaluated at t\.
3.214 Newcomb's Precession A section on Newcomb's precession angles has been
included since it is relevant for the transformation of old catalogs. Since Newcomb's
value of the precession constant (Newcomb, 1895) came into general use, several
representations have been used for the practical realization of the accumulated
precession angles denoted by Coi Z> and 9, which distinguishes them from Lieske's
precession angles C\, Za, and 9A. One may cite the formulations given in the Ex-
planatory Supplement (1961), the development by Andoyer (1911), from which the
3 / CELESTIAL REFERENCE SYSTEMS
107
Table 3.214.1
Precession Angles 1984 January 1d0h
Woolard and Clemence
Kinoshita
ES
Adopted
Co
783. 70925
783. 70938
783. 70798
783. 7092
z
783.80093
783. 80106
783. 79942
783. 8009
e
681 . 38830
681 . 38849
681.38732
681.3883
discussion by Woolard and Clemence (1966) was drawn, and the formulation given
in the introduction to the SAOC (1966), as well as a more recent discussion by Ki-
noshita (1975) and repeated by Aoki et al. (1983), where the discussion is carried
to one more significant figure than that given by Andoyer.
Among the various representations, differences in the accumulated precession
angles of the order of 1 mas are found in as short a time as 30 years. For example,
a comparison of results from Woolard and Clemence (1966), Kinoshita (1975), and
the Explanatory Supplement (ES) for the accumulated precession angles £o> Z, and
6 between equinox and equator of B1950.0 and the equinox and equator of 1984
Jan ld0h is shown in Table 3.214.1.
Results from the Kinoshita formulation agree with that of Woolard and Clem-
ence at the 0.1 mas level. Disagreement of the results from the ES at the 1 to 2 mas
level is brought about by the neglect of higher-order terms in the ES (p. 30). The
Andoyer (Woolard and Clemence) expressions for the precession angles are
Co = (23035'.'545 + 139"720f, + 0'.'060ff ) r + (30'.'240 - 0"270f1) r2 + 17'.'995 r3
z = (23035 :'545 + 139"720jr, +o:'060r|)r + (109"480 + 0'.'39QffK2 + 18'.'325t3
6 = (2005 r.'12 - 85:29 4 - 0?37 tf)r + (-42'.'65 - 0737 h)r2 - 41'.'80 r3
(3.214-1)
where f] and ti are the intervals in units of 1000 tropical years of 365242.198782
days between the initial fixed epoch eF and B1850.0, and between the final epoch
of date eo and B1850.0, respectively; thus
fl = (sr - B1850.0) / (1000 tropical years),
t2 = (sD - B1850.0) / (1000 tropical years),
and t = ti — t\.
(3.214-2)
1 08 EXPLANATORY SUPPLEMENT
The rates of general precession in right ascension mo and declination no are
given by
4607l'.'090 + 279:'440/i +0l'120/f
2005 1 r 1 2 - 85?29 r, - 0:'37 1\ (3.214-3)
3.215 Approximate Reduction for Precession and Nutation The following formu-
las are based on the day number method (see Section 3.341), and transform the
right ascension and declination (ao, ^o) from the standard epoch and equinox of
J2000.0 to the true equinox and equator of date,
a = o-q + / + g sin(G + a0) tan 60,
6 = 60+gcos(G + a0), (3.215-1)
where /, g, and G are given by
f = M + A(m/n) + E,
g = (B2 + (A+N)2)v\
C = tan-'(B/(A + A0)+ ^M. (3.215-2)
The expressions for /, g, and G are an extension of the use of the day numbers,
which apply precession and nutation from the epoch t (e.g., the middle of the nearest
year) only to the epoch of date r+r, where T is the fraction of a year. Thus A, B, and
E are the Besselian day numbers, and m and n are the annual rates of precession
(see Section 3.342). The additional terms M and N (Equation 3.213-2), which are
the accumulated precession angles, are requiredHo apply precession from J2000.0
to epoch t.
The values of /, g, and G are tabulated at 10-day dates in The Astronomical
Almanac.
no = —(9)\T=0 -
or
3.216 Differential Precession and Nutation The corrections for differential preces-
sion and nutation are given below. These are to be added to the observed difference
in the right ascension and declination, Aq and AS, of an object relative to a com-
parison star to obtain the differences in the mean place for a standard epoch (e.g.,
J2000.0 or the beginning of the year). The differences Aa and AS are in the sense
"object - comparison star," and the corrections are in the same units as Aa and
AS. In the correction to right ascension the same units must be used for Aa and AS.
3 / CELESTIAL REFERENCE SYSTEMS
correction to right ascension =
correction to declination =
where e =
/ =
109
e tan S Aa- j sec2 S AS,
/Aa,
— cos a (nt + sin e A^) - sin a Ae,
+ sin a (nt + sin e Ail') - cos a Ae,
e = 23?44, (3.216-1)
t is the time from the standard epoch to the time of observation, and n the rate
of precession in declination. If / is in years then n is in radians per year, and Axjj
and Ae are the nutations in longitude and obliquity at the time of observation,
expressed in radians.
The errors in arc units caused by using these formulas are of order lO-8^ sec2 <5
multiplied by the displacement in arc from the companion star.
3.22 Nutation
The long-period motion of the Earth's rotation axis with respect to the axis of
the ecliptic caused by lunisolar torque is call lunisolar precession. The short-period
motion of the Earth's rotation axis with respect to a space-fixed coordinate sys-
tem is called nutation. It is intimately connected with polar motion, which is the
movement of the Earth's rotation axis with respect to an Earth-fixed coordinate
system. This short-period rotational motion of the Earth includes effects of both
forced motion and free motion. The forced motion is due to the torque produced by
the gravitational attraction of the Moon and to a lesser extent that of the Sun and
planets on the equatorial bulge of the Earth and any deformations of the Earth.
For the free motion, the external forces are set to zero in the equations of motion,
which then yield particular integrals. The free motion can be determined only from
observations and may be excited by internal processes. If the Earth were a rigid
body, it would be possible to compute the coefficients of the forced periodic terms.
The observed values of some of the terms are different from the calculated values,
because of the difficulty of modeling the nonrigidity of the Earth. The principal
difference occurs in the coefficient of the 18.6-year nutation terms, which is known
as the constant of nutation.
The reference pole for nutation and polar motion is called the Celestial Ephem-
eris Pole. It is chosen to be along the axis of figure for the mean surface of a model
of Earth in which the free motion has zero amplitude. This pole has the advan-
tage that it has no nearly diurnal components of motion with respect to either
a space-fixed or an Earth-fixed coordinate system. Previously, the ephemeris pole
was chosen to be the instantaneous axis of rotation, but it was pointed out by
Atkinson (1973, 1975) that this led to quasi-diurnal motions and was not the best
reference pole.
1 1 0 EXPLANATORY SUPPLEMENT
The Celestial Ephemeris Pole is the axis about which the diurnal rotation of
the Earth is applied in the transformation between celestial and terrestrial frames,
and so it is often referred to as the axis of rotation in general discussions, as it was
in the first paragraph of this section. It may also be referred to as the true celestial
pole of date (IAU, 1976).
3.221 Celestial Ephemeris Pole The principal axes of a body are the orthogonal
directions that define a cartesian coordinate system in which the moment of inertia
tensor is a diagonal matrix, i.e., the products of inertia are zero. For a rigid ellipsoid
of revolution, two of the principal moments of inertia are equal, and the principal
moment is larger about the third. This latter axis is the axis of figure. The Earth
imperfectly resembles an ellipsoid of revolution, and, in addition, is elastic. Thus
its inertial tensor, and the directions of the principal axes, are functions of time.
Nevertheless, the instantaneous axis of figure F can be defined as the line passing
through the center of mass of the Earth, which is parallel to the primary eigenvector
of the instantaneous inertia tensor of the Earth. The intersection of F with the
surface of the Earth is the instantaneous pole of figure.
The instantaneous axis of figure of a deformabie Earth is subject to substantial
motions due to distortions of the Earth, such as those caused by the body tides.
This effect makes it difficult to interpret the motions of axis F, above, in terms of
intuitive, rigid-body concepts. It then becomes useful to define the mean surface
geographic axis, B, as an axis attached in a least-squares sense to the Earth's outer
surface. Consider a network of observatories on a rigid Earth and an axis fixed
with respect to the positions of the observatories. The only possible motion of the
observatory network is a possibly time- dependent, rigid rotation; that portion of
the rigid rotation which is not parallel to the axis will cause motion of the axis as
the latter follows motions of the observatories defining it. On a deformabie planet
we must allow for the possibility of other motions of the observatories, and we
generalize the definition of the axis so that it is defined in a least-squares sense by
the position of the observatories. The axis B is exactly this axis in the limit of an
infinite number of uniformly distributed observatories. Thus if we decompose the
motion of the Earth's surface into a mean rigid rotation plus a residual deformation,
B moves with the sense prescribed by the mean rotation. The axis B does not
respond to body tides. Note that for a rigid Earth B coincides with F at all times.
There is no nontidal periodic motion of observatories on the Earth with respect to
B. There are motions of observatories with respect to B caused by crustal motions
and, of course, by tidal forces.
The Conventional International Origin, the CIO, was defined by adopted lati-
tudes of the five International Latitude Service (ILS) observatories contributing to
the International Polar Motion Service (IPMS) (Markowitz and Guinot, 1968). It
is not known whether the CIO is fixed with respect to B, but it is believed that the
CIO roughly coincides with the mean pole F over the period 1900 to 1905. Later the
3 / CELESTIAL REFERENCE SYSTEMS
111
terrestrial pole was defined by a larger number of observatories operating optical
equipment (e.g., PZT's). This system attempted to maintain continuity with the
CIO system. It was coordinated by the Bureau International de l'Heure (BIH), and
so it is called the BIH pole. More recently SLR and VLBI have become the most
accurate observational techniques, and the BIH pole became strongly dependent
on these data. In 1988 January the international activities of timekeeping and of
monitoring Earth orientation were separated, and the maintenance of the terrestrial
reference frame was taken over by the International Earth Rotation Service (IERS),
again attempting to maintain the same pole as the CIO and BIH, but because of the
higher accuracies now obtained it is necessary to include a model of plate motions
in the reduction of observations. This reference frame is called the IERS Terres-
trial Reference Frame (ITRF). International timekeeping is now coordinated by the
Bureau International des Poids et Mesures (BIPM).
The terms nutation and polar motion have been defined as the forced and free
motions, respectively, of the adopted reference pole (the Celestial Ephemeris Pole),
but in some cases the separation of these effects is not clear. The forced solution is
the solution of the Earth's equations of rotational motion that accounts for all the
external forces (specifically the gravitational forces due to the Sun, Moon, and plan-
ets). The free solution is that solution which results from setting the external forces
to zero (the particular solution of the differential equations) . For a rigid Earth this
results in the Eulerian free motion, a component of polar motion, which would have
annual periods in a terrestrial frame. For a nonrigid Earth the period is lengthened
to about 14 months, and is called the Chandler component of polar motion. Hence
the forced annular polar motion due to the climate along with motions resulting
from geophysical strains and meteorological effects should logically be included in
nutation, but in practice they are included in polar motion because these effects
cannot be calculated as accurately as the external forcing effects.
For an Earth with a fluid core there exist a second free solution, the so-called
nearly diurnal free polar motion (also called the nearly diurnal free wobble or the
free core nutation), which has not yet been clearly observed. Like polar motion it
is part of the transformation from space-fixed to body-fixed coordinates (Yatskiv,
1980).
3.222 The 1980 IAU Theory of Nutation and Reduction The 1980 IAU Theory of
Nutation was computed for the Celestial Ephemeris Pole by determining the nu-
tations in longitude and obliquity of a rigid Earth (Kinoshita, 1977), and making
modifications for the nonrigid Earth model 1066 A (Gilbert and Dzieiwonski, 1975)
by Wahr (1981) in such a fashion that there are no nearly diurnal motions of this
celestial pole with respect to either space-fixed or body-fixed (crust-fixed) coordi-
nates. Tables 3.222.1 and 3.222.2 give the 1980 IAU Nutation series and the set of
fundamental arguments (Van Flandern, 1981) based on the orbital motion of the
Earth and Moon needed for evaluating the series.
Table 3.222.1
Nutation in Longitude and Obliquity Referred to the Mean Ecliptic of Date
Argument Coefficient for
Multiple of Period Longitude Obliquity
No.
/
1
Si
a
k
F
D
Q
Bt
fdays)
Si
1" * 10
-4
1
0
o
6798.4
— 1 7 1 996
— 174.2 7
92025
+8.9 7
2
rt
K)
o
2
2062
+0.2 7
— 895
+0.5 7
3
— 2
0
2
0
1
1305.5
46
24
4
2
0
-2
0
0
1095.2
11
0
5
-2
0
2
0
2
1615.7
-3
1
6
1
— 1
0
—3
0
7
0
— 2
z
— 2
1
6786.3
—2
1
Q
2
0
—2
o
1
943.2
•|
g
g
0
0
2
-2
2
182.6
-13187
-1.67
5736
-3.1 7
10
0
1
0
0
0
365.3
1426
—3.4 T
54
-0.1 7
1 1
0
1
2
— 2
2
121 .7
—517
+1.2 7
224
-0.6 7
12
0
— 1
2
—2
2
365.2
217
—0.5 7
-95
+0.3 T
13
o
o
2
_2
177 8
129
+0 1 7
— 70
14
2
0
0
-2
0
205.9
48
1
15
0
0
2
-2
0
173.3
-22
0
16
0
2
0
0
0
182.6
17
—0.1 7
0
17
0
1
0
0
1
386.0
- 15
9
1B
o
2
2
_2
2
91.3
+0 1 7
ig
0
-1
0
0
"1
346.6
-12
6
20
-2
0
0
2
1
199.8
-6
3
21
0
— 1
2
— 2
1
346.6
—5
3
22
2
0
0
—2
1
212.3
4
-2
23
0
2
_2
■J
1 19 6
4
— 2
24
1
0
0
-1
0
411.8
-4
0
25
2
' T
0
-2
0
131.7
1
0
26
0
— 2
2
1
169.0
0
jj
1
— 2
2
0
329.8
0
28
0
o
2
409 2
0
29
-1
0
0
1
1
38B.3
g
30
0
1
2
-2
0
117.5
0
31
g
y
2
0
2
13.7
—2274
—0.2 7
977
-0.5 7
32
0
0
0
0
27.6
712
+0.1 7
—7
33
0
o
2
o
■|
13 6
-0.4 7
200
34
1
0
2
0
2
9.1
-301
129
—0.1 7
35
1
0
0
-2
0
31.8
-158
_1
36
, ^
0
2
0
2
27.1
123
-53
37
U
0
0
2
0
14.6
63
-2
38
1
o
o
o
■|
£7 7
63
+0. 1 7
—33
39
-1
0
0
0
1
27.4
—58
—0.1 7
32
40
-1
0
2
2
2
9.6
-59
26
41
1
0
2
0
1
9.1
-51
27
42
0
2
2
2
7.1
-38
16
43
2
o
0
0
13 8
29
— 1
44
1
0
2
-2
2
23.9
29
— 12
45
2
0
2
0
2
6.9
—31
13
46
0
0
2
0
0
13.6
26
-1
47
■1-1
0
2
0
1
27.0
21
-10
48
0
0
2
1
32.0
16
-B
49
1
0
0
-2
1
31.7
-13
7
50
-1
0
2
2
1
9.5
-10
5
51
1
1
0
-2
0
34.8
-7
0
52
0
1
2
0
2
13.2
7
-3
53
0
-1
2
0
2
14.2
—7
3
54
1
0
2
2
2
5.6
-8
3
55
1
0
0
2
0
9.6
6
0
3 / CELESTIAL REFERENCE SYSTEMS
113
Table 3.222.1, continued
Nutation in Longitude and Obliquity Referred to the Mean Ecliptic of Date
Argument
Coefficient lor
Multiple of
Period
Longitude Obliquity
No.
/
F
D
Si d
i
3j
hi
Ql
(days)
r x io-4
56
2
0
2
-
.
6 —3
57
0
0
0
2
1
14 B
g 3
58
q
o
2
2
1
7,1
-7 3
59
1
o
2
-2
1
23.9
6 -3
60
0
0
0
-2
1
14.7
-5 3
61
1
—1
o
0
0
29.8
ft ft
62
2
o
2
0
1
6.9
— D o
63
1
0
_2
0
15.4
-4 0
64
1
o
—2
0
0
26.9
4 0
65
0
0
0
■j
0
29.5
-4 0
66
1
1
0
0
0
25.6
— J u
67
1
o
2
0
0
9.1
68
"I
—1
2
0
2
9.4
-3 1
69
-1
—1
2
2
2
9.8
-3 1
70
-2
0
0
0
1
13.7
-2 1
71
3
o
2
0
2
Q 1
72
0
_1
2
2
2
7.2
— o I
73
■I
1
2
0
2
8.9
2 -1
74
-1
o
2
-2
1
32.6
-2 1
75
2
0
0
0
1
13.8
2 -1
76
1
o
o
0
2
27.8
—2 1
77
3
o
o
0
0
9.2
78
q
o
2
1
2
9.3
2 -1
79
-1
o
o
0
2
27.3
1 -1
ao
1
0
0
-4
0
10.1
-1 0
81
2
o
2
2
2
14.6
1 — i
82
— 1
o
2
4
2
5.8
— e. I
83
2
o
o
_4
0
15.9
-1 0
84
1
1
2
-2
2
22.5
1 -1
85
1
0
2
2
1
5.6
-1 1
86
—2
o
2
4
2
7.3
— 1 1
S7
— 1
o
4
0
2
9.1
1 0
88
i
o
_2
0
29.3
1 ' If
89
2
o
2
-2
1
12.8
1 -1
90
2
0
2
2
2
4.7
-1 0
91
1
o
o
2
1
9.6
4 Q
92
0
q
4
— 2
2
12.7
1 ft
93
o
2
2
2
8.7
, in' o
94
1
o
2
-2
0
23.8
-1 0
95
0
1
2
0
1
13.1
1 0
96
'
— 1
o
2
1
35,0
1 0
97
0
0
0
1
13.6
— 1 0
98
o
0
2
_ 1
2
25.4
— 1 0
99
0
1
0
2
0
14.2
-1 0
100
1
0
-2
-2
0
9.5
-1 o
101
0
-1
2
0
1
14.2
-1 0
102
1
1
0
-2
1
34.7
-1 o
103
1
0
-2
2
0
32.8
-1 0
104
2
0
0
2
0
7.1
1 0
105
0
0
2
4
2
4.8
-1 0
106
0
1
0
1
0
27.3
1 0
114
EXPLANATORY SUPPLEMENT
Table 3.222.2
Fundamental Arguments
/ = 134°57'46'.,733 + (1325r + 1 98° 52' 02"633) T + 311310 T2 + Ql'064 7"3
P = 357°31'391'804 + (99r + 359° 03' 01 l'224)T - 0'.'577 T2 - 0:'012 T3
F = 93° 1 6' 1 8"877 + (1 342r + 82° 0 V 03!' 1 37) T - 1 3?257 T2 + 01'01 1 T3
D = 297°5r01"307 + {1236r + 307°06'4i:'328)r - 61'891 T2 + 01'01 9 T3
Q = 135°02'401'280 - {5r + 134°08' 101'539)7" + 7:'455 T2 + O'.'OOS 7"3
Note: 1 ' = 360° and T => (JD - 2451545. 0) / 36525 is the number of Julian centuries of 36525 days
of 86400s of dynamical time from the fundamental epoch of J2000.0 to date. The fundamental
arguments are:
/ = the mean longitude of the Moon minus the mean longitude of the Moon's perigee,
/' = the mean longitude of the Sun minus the mean longitude of the Sun's perigee,
F = the mean longitude of the Moon minus the mean longitude of the Moon's node.
D = the mean longitude of the Moon minus the mean longitude of the Sun. i.e., the mean
elongation of the Moon from the Sun,
f> - the longitude of the mean ascending node of the lunar orbit on the ecliptic measured Jrom
the mean equinox of date.
Nutation describes the motion of the true pole relative to the mean pole and
may be resolved into the components Aip in longitude and Ae in obliquity, which
are shown in Figure 3.222.1, The nutation matrix N is a sequence of three rotations
which uses these angles and the mean obliquity of the ecliptic, eo, to transform
equatorial coordinates referred to the mean equinox and equator of date to the true
equinox and equator of date.
The rotations are a rotation of eg about the jr-axis, which transforms the
Ay-plane from the mean equator of date to the mean ecliptic of date (see Fig-
ure 3.222.1), followed by a rotation of — Aip about the z-axis, which applies nutation
in longitude and rotates the jr-axis, pointing toward the mean equinox of date T,
to the true equinox of date TT, and lastly a rotation of — e, the true obliquity of
date, about the *-axis, which rotates the xy-plane to the true equator of date. The
expressions for the mean and true obliquity of date are given by
e0 = 23°26'2r.'448 - 46'.'8150 T - 0'.'00059 + 0"001813 7"3, (3.222-1)
e = e0 + Ae, (3.222-2)
where T = (3D — 245 1545. 0) / 36525. The nutation matrix is
N = Ri(-e)R3(-A#Rl(+eo)) (3.222-3)
3 f CELESTIAL REFERENCE SYSTEMS
115
Mean ecliptic of date
Mean equator of date
True equator of date
Figure 3.222.1
The mean and true equators of date
which is rewritten in full as
■ cos Aip - sin Aip cos e0 - sin Aip sin e0
sin Aip cose cos A^ cos e cos en + sin e sin en cos Aip cose sin en — sine cos eo
- sin A^' sin e cos At/' sin e cos eo - cos e sin e0 cos Aip sin e sin e0 + cos e cos eo -
(3.222-4)
Nutation may be rigorously applied to equatorial rectangular coordinates thus:
N =
r = Nr0,
(3.222-5)
where the position vector Tq referred to the mean equinox of date is transformed to
the position vector r referred to the true equinox of date.
The nutation matrix is taken from Emerson (1973), and it is the complete
rotation matrix without any approximations. The nutation angles Aip, Ae may be
evaluated from
Aip = Sj sin A,- and Ae = ^ C, cos A,
i=i
where A, = a;/ + £>,/' + c,F + d,D + etQ
(3.222-6)
and the multipliers Of, tti, c,, dh e„ and the coefficients 5, and C, are given in
Table 3.222.1, which defines the 106 terms of the IAU (1980) nutation series. The
terms of the series are grouped according to their periods, and arranged roughly
in order of magnitude of the coefficient of the nutation in longitude within each
group. The fundamental arguments /, /', F, D, and f2, in the FK5 reference system,
are given in Table 3.222.2. Values for these nutation angles and e, calculated using
these series, are tabulated daily in The Astronomical Almanac.
116
EXPLANATORY SUPPLEMENT
3.223 The Equation of the Equinoxes and the Uniform Equinox The equation of
the equinoxes, which in editions prior to 1960 was called "nutation in right ascen-
sion," is the right ascension of the mean equinox referred to the true equator and
equinox and equals A^cos e. In Figure 3.222.1 it is represented by the arc TTQ and
it represents the difference between the true and mean right ascensions of a body
on the true equator; it is thus the difference between apparent and mean sidereal
time; i.e.,
GAST = GMST + Ay cos e; (3,223-1)
where GAST and GMST are the Greenwich apparent and mean sidereal times,
respectively (Section 2.23), which, together with the equation of the equinoxes are
tabulated daily to OfOOOl in The Astronomical Almanac.
It is sometimes more convenient to refer right ascensions to the uniform equinox
[Q in Figure 3.222.1) (Atkinson and Sadler, 1951), which is the true equinox of date
minus the equation of the equinoxes. The advantages are that in the determina-
tion of hour angle, right ascensions of objects referred to the uniform equinox are
subtracted from mean sidereal time. These right ascensions are smoother functions
of time which reduce the number of terms required when they are represented by
polynomial or Fourier-type expression over a fixed time interval. It also simplifies
interpolation if tables give mean sidereal time and right ascension referred to the
uniform equinox.
3.224 More Accurate Nutation Recent analysis of highly accurate observations
(VLBI) indicates the need for some systematic corrections to the IAU 1980 Nuta-
tion series. Presumably these corrections are due to the difference of the real Earth
obtained from the model of Wahr. The latest differences Herring (1987) published
in BIH Annual Report for 1987 (1988) are given in Table 3.224.1. The corrections
Table 3.224.1
Corrections to IAU 1980 Nutation Series
Argument
Coefficient for
Multiple of
Period
Longitude Obliquity
No.
t
r
F
D
f?
LSn
LCn OCn OSn
n
an
br
dn
e„
days
1"x 10-5
1
0
0
0
0
1
6798.4
-725
417 213 224
2
0
1
0
0
0
365.3
523
61 208 -24
3
0
0
2
-2
2
182.6
102
-118 -41 -47
4
0
0
2
0
2
13.7
-81
0 32 0
3 / CELESTIAL REFERENCE SYSTEMS
117
Arpc and Aec to be added to the IAU 1980 nutations in longitude and obliquity,
Atj} and Ae, respectively are
4
At/;c = Mn4« + LC„ cos A„),
4
Aec = ^2 (°c» cosA" + 0S" sin A»)> (3.224- 1 )
n=l
where A„, the sum of the required arguments is given in Equation 3.222-6. These
corrections are made up of two terms: the first terms, LS„ for longitude and OC„
for obliquity are called in-phase, or direct: the other two terms LC„ and OS„, are
called out-of-phase, or indirect.
There are other effects due to the planets that are called the planetary terms
in nutation. Vondrak (1983) has produced a table (Table 3.224.2) of 85 terms that
gives the combined direct and indirect planetary effects that have a maximum
amplitude of several parts in 10~4 arcseconds and should be used if inilliarcsecond
accuracy is required.
The planetary nutation in longitude, Aipp, and in obliquity Aep may be calcu-
lated from
85
Aij}p - ^2 sinA,, + LC„ cos A„)
n=l
85
Aep = (°C>> cos A» + 0S" sin An) (3.224-2)
rf=1
and A„ = a„ I + b„F + c„D + dnQ + enQ + f„V + gnE + h„M + inJ +j„S
where LS„, LC„, OC„, OS,,, and the multipliers
ani bm cni /ni £ni '«i find jn
are given in Table 3.224.2. The fundamental arguments I, F, D, and J? are given in
Table 3.222.2. Q, V, E, M, J, and 5, the mean heliocentric longitudes of the planets
Mercury, Venus, Earth, Mars, Jupiter, and Saturn, respectively, are given as:
Q =252? 3+ 149472.7 7 M= 353?3 + 19140. 3 T
V= 179?9 + 58517. 8 T J= 32?3 + 3034.9 7 (3.224-3)
£=98?4 + 35999. 4 7 S= 48?0 + 1222. 1 7
where T is measured in Julian centuries from 2451545.0.
The IAU is at present studying various proposals prior to adopting an improved
nutation series. In the future there will be a need for an improved nutation transfor-
mation with improvements to both the rigid body and nonrigid body components.
Table 3.224,2
Planetary Terms in Nutation, Combined Direct and Indirect Effects
Argument
Multiple of
Period
Coefficient for
Longitude Obliquity
No.
n
/
F
bn
D
Cn
a
dn
G
en
V
fn
E
9n
m
hn
J
in
s
in
days
LSn
LCn
r x
OCn
10-5
OSn
1
0
n
u
Q
0
0
o
2
Q
o
0
1 83
_1
o
o
o
2
1
0
-2
-1
3
0
-1
0
0
0
12924
0
-1
0
0
3
0
0
0
0
0
1
-2
0
0
0
975
0
-1
0
0
4
0
0
0
0
0
1
-1
0
0
0
584
16
0
0
0
5
0
0
0
0
0
1
1
0
0
0
139
-4
0
0
1
6
0
u
u
n
u
n
i
— o
n
yj
ft
u
ft
y
_-|
q
o
o
7
0
0
0
0
0
2
-4
0
0
0
488
4
0
0
3
8
0
0
0
0
0
2
-3
0
0
0
1455
0
7
0
0
9
0
0
0
0
0
2
-5
0
0
0
209
0
-1
1
0
10
0
0
0
0
0
2
-2
0
0
0
292
-6
t
0
0
1 1
u
n
u
f\
U
n
U
C.
_ 1
u
ft
o
162
o
2
_-l
Q
12
0
0
0
0
0
2
0
0
0
0
112
4
0
0
-2
13
0
0
0
0
0
3
-7
0
0
0
172
1
0
0
0
14
0
0
0
0
0
3
-6
0
0
0
325
0
-1
0
0
15
0
0
0
0
0
3
-5
0
0
0
2959
20
-4
1
7
1 b
u
n
u
n
U
I
n
U
— o
£
0
0
o
5240
q
•|
o
o
17
0
0
0
0
0
3
-4
0
0
0
417
0
4
0
0
18
0
0
0
0
0
3
-3
0
0
0
195
2
0
0
0
19
0
2
-2
1
0
-3
3
0
0
0
2061
1
0
0
-1
20
2
0
-2
0
0
-3
3
0
0
0
3562
1
0
0
0
u
U
U
u
n
u
(9
u
fl
u
Q
127
o
_1
o
22
0
0
0
0
0
4
-7
0
0
0
734
0
1
0
0
23
0
0
0
0
0
4
-6
0
0
0
728
-5
1
-1
-2
24
0
0
0
0
0
4
-4
0
0
0
146
1
0
0
0
25
0
0
0
0
0
5
-8
0
0
0
2864
0
-3
2
0
to
n
U
ft
u
ft
y
ft
u
U
e
a
— f
u
n
u
2
0
n
I
27
0
2
-2
1
0
-5
6
0
0
0
4948
0
-4
-2
0
28
0
2
-2
0
0
-5
6
0
0
0
18186
0
2
0
0
29
0
0
0
0
0
5
-5
0
0
0
117
-1
0
0
0
30
0
0
0
0
0
6
-9
0
0
0
485
0
-1
0
0
n
U
U
n
U
u
n
u
c
D
n
u
n
u
u
1
**" 1
n
ft
_1
32
2
0
2
0
0
-6
8
0
0
0
17496
-1
0
0
0
33
0
0
0
0
0
7
-9
0
0
0
154
1
0
0
0
34
0
0
0
0
0
8
-10
0
0
0
122
-1
0
0
0
35
0
0
0
0
0
8
-15
0
0
0
183
-1
-1
0
0
ob
0
0
0
0
0
8
—13
0
0
0
87265
3
— 4
V
ft
u
Jf
0
0
0
1
0
8
—13
0
0
0'
7373
1
1
u
n
u
38
0
0
0
i .
0
-8
13
0
0
0
6307
-1
1
0
0
39
0
0
0
0
0
8
-11
0
0
0
182
-1
-1
0
0
40
1
0
0
-1
0
-10
3
0
0
0
6733
0
-1
0
0
41
0
0
1
0
-10
3
0
0
0
6665
0
-1
0
0
42
0
0
-1
0
-18
16
0
0
0
7296
1
0
0
0
43
0
0
1
0
-18
16
0
0
0
6364
1
0
0
0
44
-2
0
-2
0
-18
16
0
0
0
14
1
-1
0
1
45
2
0
2
0
-18
16
0
0
0
14
1
-1
0
-1
3 / CELESTIAL REFERENCE SYSTEMS
119
Table 3.224.2, continued
Planetary Terms in Nutation, Combined Direct and Indirect Effects
Argument
Multiple of
Coefficient for
Period Longitude Obliquity
No.
/
F
D
S2
o
V
e
M
J
S
LSn
LCn
OCn
osns
n
3n
K
On
Cn
fir.
J n
yn
hn
in
jn
days
T x
-«— 5
46
i
0
-2
1
0
20
-21
0
0
0
14610
0
1
1
0
47
0
0
0
0
0
0
1
-1
0
0
780
-1
0
0
0
48
0
0
0
0
0
0
1
2
0
0
177
-1
-1
0
0
49
0
0
0
0
0
0
1
-2
0
0
5764
-5
3
0
0
50
0
0
0
1
0
0
-1
2
0
0
37883
-6
-6
-3
3
51
0
0
0
0
0
0
0
2
0
0
344
-1
0
0
0
52
0
0
0
0
0
0
2
-2
0
0
390
-4
0
0
0
53
0
0
0
0
0
0
3
-2
0
0
189
1
-1
0
0
54
0
0
0
0
0
0
4
-2
0
0
124
1
o
o
-1
55
0
0
0
0
0
0
2
-3
0
0
902
-1
0
0
0
56
0
0
0
0
0
0
2
-4
0
0
2882
0
1
0
0
57
0
0
0
0
0
0
3
-4
0
0
418
-1
1
0
0
58
0
0
0
0
0
0
2
-8
3
0
183
1
-4
2
0
59
0
0
0
1
0
0
-4
8
-3
0
6870
1
4
2
-1
60
0
0
0
1
0
0
4
-8
3
0
6720
-1
3
2
1
61
0
0
0
0
0
0
6
-8
3
0
183
-4
-2
-1
62
0
0
0
0
0
0
1
0
-3
0
489
1
0
0
0
63
0
0
0
0
0
0
2
0
-3
0
209
1
0
0
0
64
2
0
-2
-1
0
0
-2
0
3
0
4528
1
0
0
1
65
2
0
-2
1
0
0
-2
0
3
0
13630
2
0
0
-1
66
2
0
-2
0
0
0
-2
0
3
0
13563
4
0
0
0
67
0
0
0
0
0
0
1
0
-2
0
439
-2
3
0
0
68
0
0
0
0
0
0
2
0
-2
0
199
4
0
0
0
69
2
0
-2
-1
0
0
-2
0
2
0
100130
-5
0
0
-3
70
2
0
-2
1
0
0
-2
0
2
0
3288
1
0
0
0
71
0
2
-2
1
0
0
-2
0
2
0
1642
1
0
0
0
72
2
0
-2
0
0
0
-2
0
2
0
6366
4
0
0
0
73
0
0
0
0
0
0
3
0
-2
0
129
1
-1
0
0
74
0
0
0
0
0
0
4
0
-2
0
95
-2
0
0
1
75
0
0
0
0
0
0
1
0
-1
0
399
-16
0
0
0
76
0
0
0
0
0
0
2
0
0
191
-2
0
0
1
77
0
0
0
0
0
0
3
0
0
125
5
1
0
2
78
0
0
0
0
0
0
0
0
0
4333
-1
-1
-1
0
79
0
0
0
1
0
0
0
0
0
11945
0
-1
0
0
80
0
0
0
0
0
0
1
0
0
337
-1
0
0
1
81
0
0
0
0
0
0
2
0
1
0
175
2
0
0
-1
82
0
0
0
0
0
0
0
0
2
0
2166
-12
2
1
0
4
83
0
0
0
0
0
0
1
0
2
0
313
0
-1
0
84
0
0
0
0
0
0
0
0
3
0
1444
-1
0
0
0
85
0
0
0
0
0
0
1
0
0
-1
378
-1
0
0
0
Explanatory Supplement to The Astronomical Almanac, seventh proof, July 3tf, 1992
1 20 EXPLANATORY SUPPLEMENT
3.225 Approximate Nutation and Reduction The nutation matrix given in Equa-
tion 3.222-4 can be simplified to
N =
l -A^cose -A^sine
+ Accost 1 -Ae
+A^sine +Ae ]
(3.225-1)
where Aip and Ae are expressed in radians and second-order terms have been
neglected, since they can reach only one unit in the eighth decimal place.
Equatorial rectangular coordinates referred to the mean equinox can be con-
verted to the true equinox by application of the corrections
Ax - -{ycos £ + zsin e)A^,
Ay - +jccos eAip - zAe,
Az = +xsineAi£> + yAe. (3.225-2)
Similarly the first-order corrections (Aa, AS) to right ascension and declination
may be calculated directly from
Aa = (cos e + sin e sin a tan 6)A tl> - cos a tan SAe,
AS = sine cos aA4> + sin aAe. (3.225-3)
These can be combined with the reduction for precession from the mean equinox
of the middle of the year by means of day numbers (see Sections 3.215 and 3.34).
It is also possible to calculate the nutations in longitude and obliquity to about
1" by considering only the dominant terms of the series (see Table 3.222.1), thus
A^> - -0°0048 sin(125?0 - O?05295rf) - 0?0004sin (200?9 + l°97129rf)
Ae = +0?0026 cos (125?0 - 0?05295i/) + 0?0002 cos (200?9 + 1 ?97 129rf),
(3.225-4)
where d is the number of days from 2451545.0.
3.226 Differential Nutation For objects within a small area of sky differential nu-
tation is always combined with differential precession; see Section 3.216.
3 / CELESTIAL REFERENCE SYSTEMS
3.23 Space Motion
121
The motion in space of stellar objects as a function of time must be considered since
star catalogs give the position of a star at an epoch referred to a mean equator and
equinox at a chosen epoch.
Traditionally, stellar motion has been divided into proper motion, which occurs
perpendicular to the line of sight and is given as components in right ascension (pa)
and declination {p&), and radial velocity r, which occurs along the line of sight. In
fundamental catalogs the epoch of the star positions is identical with that of the
equinox to which the catalog refers, but in observational catalogs the epoch may
differ from that of the equinox to which the positions are referred. In all cases,
the correction to position consists of the product of the star's space motion vector
with the interval between the required epoch and the epoch of the star's position
in the catalog. In some catalogs the secular variations of the proper motions are
also included, and in such cases the mean value of the proper motion during the
interval must be used. For double stars the orbital motion of the components with
respect to each other may also have to be included.
The position vector r of a star at the catalog epoch t is given by
r =
/-cos 6 cos q
rcos S sin a
rsin6
(3.23-1)
where a and 6 are the catalog right ascension and declination, and r is the barycen-
tric distance to the star in AU, which can be computed from
;• = 1 / sin p,
(3.23-2)
where p is the parallax of the star. If p is unknown, unavailable, or zero to within
the accuracy of measurement, set it to some small but finite positive number in
order to avoid mathematical indeterminacy; a choice of 1" x 10-7 will effectively
place objects of unknown parallax at a radius of lOMpc.
The space motion vector r can be obtained by differentiating Equation 3.23-1,
giving
— cos 6 sin a
— sin S cos a
cos 6 cos a '
1 5srpa
r -
cos 6 cos a
— sin S sin a
cos 8 sin a
sr/is
0
cos S
sin S
kr
(3.23-3)
where pa and pf, are the proper motions in right ascension and declination, respec-
tively, and r is the radial velocity. The factors \5s, s, and k convert fia, ps, and r
into the required units for the space motion vector, assuming that pa is in units
of time. If r is required in AU per day, and ^a and p& are in seconds of time and
1 22 EXPLANATORY SUPPLEMENT
seconds of arc per Julian century, respectively, and r is in kilometers per second,
then
s = 271"/ (360 x 3600 x 36525) converts from arcseconds/cy to radians/day,
k = 86400/ (1.49597870 x 10+8) converts from km/s to AU/day. (3.23-4)
All catalogs do not necessarily give the proper motions as rates per century; some-
times they are per year.
It is worth considering the cases in which the available data are incomplete. The
radial velocity is required only for relatively nearby stars for which foreshortening
effects (second-order changes in the apparent motion of the star due to the shifting
aspect of its motion) are significant. However, the radial velocity is useless in this
regard (and should be set to zero), if the parallax (distance) is unknown. If the star's
radial velocity is zero or unknown, or has been set to zero because the parallax is
not known, then the above space motion is tangent to the celestial sphere at the
star's catalog position r. Conversely, if the proper motion components are zero,
then the star has no known tangential velocity. Also, it may be that the proper
motion components are known but not the parallax. In such a case, if a "reasonable
guess" parallax value is not used, then the computed velocity components could be
greater than seems physically plausible. However, this is a computational curiosity
with no physical meaning or practical effect on the results of the calculation.
The position T\ of the star at epoch tj is calculated from
r, =r + (f, -/)r, (3.23-5)
where t\ - t is the interval between the required epoch and the epoch of the catalog.
If r is in AU/day then clearly (/] -t) must be in days. Use of the space motion vector
(Equation 3.23-3) includes both proper motion and foreshortening effects in t\.
The errors that are being made in this method of calculating space motion have
been investigated by Stumpff (1985). Basically, proper motion and radial velocity
are observables that should be converted to inertial quantities. Thus catalog proper
motions include the effect of aberration, whereas radial velocities include a rela-
tivists Doppler effect. Fortunately, these effects are very small and are important
for only a few nearby stars whose apparent motions are changing rapidly across
the line of sight. Stumpff has shown that the errors in neglecting these effects can
increase progressively with time, so that after several decades they may become
significant at the mas level.
There is no explicit correction for light-time in stellar apparent place compu-
tation; it is assumed that the position and space motion vectors implicitly include
the light-time and its time derivative.
3 / CELESTIAL REFERENCE SYSTEMS
123
A linear approximation to a star's motion in space, which ignores the changes
in proper motion due to the changes in right ascension and declination, and ignores
foreshortening and radial velocity, may be made when the highest accuracy is not
needed, or if the time interval (r) is small such as in the day-numbers method
(Section 3.34) when r is a fraction of a year. Corrections are made directly to the
right ascension and declination, via
ai=a + Tfj,a, 6l=6 + Tns, (3.23-6)
where r — ti — t is expressed in the appropriate units.
3.24 Parallax
Parallax is the term used to describe the difference in the apparent direction of
an object as seen from two different locations. For example, the Sun, Moon, and
planets are observed from the surface of the Earth, and most almanacs publish their
geocentric positions. The term is usually used in connection with a shift of origin
from the center of the Earth to the surface (diurnal parallax), or a shift from the
barycenter or heliocenter to the geocenter (annual parallax).
The following is a simple and rigorous method of making corrections for both
diurnal and annual parallax, and it is recommended that it be used in preference to
approximate methods. The position vector of a body rg, with respect to the origin
E, is given by
rE = uB - Eb, (3.24-1)
where Eb and Ub are the position vectors of the new origin and the body, respec-
tively, with respect to the old origin and referred to the same reference frame.
Approximate methods are given for calculating diurnal and annual parallax
that are useful when the appropriate data are not available; e.g., the distance of
the body is not known.
3.241 Diurnal Parallax Diurnal parallax is a shift of origin from the center of
the Earth to the surface of the Earth. Coordinates with an origin at the surface
of the Earth are called topocentric coordinates, and depend on the position of the
observer. Using Equation 3.24-1, the topocentric rectangular equatorial coordinates
of a body r' an1 given by
EXPLANATORY SUPPLEMENT
r' = r - g(f);
X =r! cos 8' cos a' = rcos6cosa - p cos 9/ cos (6 + A),
/ = r' cos 6' sin q' =rcos<5sina - pcos<f>' sin (0 + A),
z' = r'sin6' = /-sin<5 -psin^',
(3.241-1)
where r and g(f) are the geocentric position vectors of the body and the observer,
respectively, referred to the same reference frame, and
r/2=x'2+y'2+z'\ a' = tan-'(//A $ *S&rV//). (3.241-2)
The geocentric position of the observer is given here in terms of p, the distance
of the observer from the center of the Earth, the geocentric latitude 0, the longitude
A (east longitudes are positive), and the Greenwich sidereal time 9. If the position
of the body is referred to the true equinox of date, then apparent sidereal time
must be used. Section 3.244 gives the formulas for the position of the observer in
an Earth-fixed reference frame.
The topocentric hour angle {h') may be calculated from h' = 0 + A - a\ where
apparent or mean sidereal time is used as appropriate, and 9 and a' are expressed
in the same units. At upper meridian transit ti = 0.
124
i.e.
3.242 Approximate Diurnal Parallax Approximations to Equation 3.241-1 may
be made in various ways. For objects relatively close to the Earth, such as the
Moon, whose parallax is significant, it can be assumed that the Earth is spherical
by setting p = 1 Earth radii. For the more distant bodies, such as the Sun the
planets, or comets, whose parallax amounts to only a few seconds of arc, it can
be assumed that the geocentric and topocentric distance are equal; i e r* = r
Inus for such bodies the topocentric right ascension and declination (a',V) may
be approximated using the geocentric position (a, 6) to sufficient accuracy by the
hrst-order approximation
a' ~ a - 7T - cos 4>' sin h sec 8,
a
6> = 6 ~ * ^(sin 4>' cos 8 - cos <}>' cos h sin 8), (3.242- 1 )
where jr is the equatorial horizontal parallax, a the equatorial radius of the Earth
P the geocentric distance of the observer, and h the geocentric hour angle.
The solar parallax is given by
sin 7tq = a I A and tt0 = 8"794 1 48.
(3.242-2)
3 / CELESTIAL REFERENCE SYSTEMS -[ 25
where a and A (the unit dist ance) are defined by the IAU (1976) system of constants.
The equatorial horizontal parallax (horizontal parallax) of any object may be
calculated from
t^-kqIA (3.242-3)
where A is the geocentric distance of the object in AU. In preliminary work on
comets and minor planets, where the geocentric distance is unknown, it is convenient
to calculate parallax factors pa and pf, for each observation; these may be used, once
the geocentric distances are determined, to give the parallax corrections in the form
Aa=paI A, AS = ps J A, (3.242^1)
where Act = a - a' and A<5 = 8 - 8'. The parallax factors are calculated from
p» - 8!'794148p cos <?!>' sin ft sec <5 = 0!5862765 p cos <f>' sin h sec 8 ,
ps = 8"794 1 48 p (sin <p' cos 8 - cos <j>' cos h sin 8), (3.242-5)
where p is expressed in AU.
The parallax correction for altitude, called parallax in altitude, is given approxi-
mately by 7r cos a', where a' is the topocentric altitude: it is added to the topocentric
altitude to form the geocentric altitude to low precision (about O'.l); there is no cor-
rection for azimuth (Z). This approximation does not allow for the oblateness of
the Earth, which, for the Moon, can produce an error of up to 0'.2. For an observer
at latitude </> the oblateness of the Earth can be corrected for by adding the next
term of the expansion, which is -7r/sin2 <?icosa' + 7r/sin 2<£cosZsina', where / is
the flattening (see Section 3.244).
3.243 Approximate Annual Parallax The reduction for annual parallax from the
bary-centric place (a, 6) to the geocentric place (au 8t) is given by
cti = q + n(Xsin a - Ycosa)/ (15cos<5),
<5i = 8 + 7r(Xcosasin6 + Ksinasin<5 - Zcos<5), (3.243-1)
where it is the parallax in arcsec, X, Y, Z are the barycentric coordinates of the
Earth in AU, and the right ascension is expressed in units of time.
These expressions may be simplified by using the star constants c, d, c' , d' (see
Section 3.342) incorporated into the day-number method (Section 3.341) in order
to correct for annual parallax. Thus
ot\ — q = A a = ir(dX - cY),
126
EXPLANATORY SUPPLEMENT
Si - i =,A5 #» w(d'X - c'Y), (3.243-2)
where Act and AS are in the same units as w. Also, they may be combined with
the correction for annual aberration (see The Astronomical Almanac, B22).
The times of reception of periodic phenomena, such as pulsar signals, may be
reduced to a common origin at the barycenter by adding the light- time correspond-
ing to the component of the Earth's position vector along the direction of the object,
that is, by adding to the observed times (X cos a cos 6+Fsinttcos6+Zsin<5)/c, where
c = 173. 14 AU/day, and the light-time for 1 AU is 1 / c = 0?0057755.
3.244 Terrestrial Coordinates There are three commonly used ways of express-
ing terrestrial coordinates (i.e., Earth-fixed): (i) geocentric equatorial rectangular
coordinates, x, y, z; (ii) geocentric longitude, latitude, and distance, A, 0', p; (iii)
geodetic longitude, latitude, and height; A, <j>, h. Geodetic coordinates are referred
to a reference spheroid (an ellipse of revolution), which is normally geocentric, and
is defined by its equatorial radius (a) and flattening (/). For example, those of
MERIT (1983) are
a = 6378. 137 km and /= 1/298.257. (3.244-1)
The geodetic and geocentric longitudes of a point are the same. The following
relationships hold between the geocentric and geodetic (geographic) coordinates:
x = p cos 4>' cos A = (aC + /z)cos0cosA,
y = pcos0'sinA = (aC + h) cos 0 sin A, (3.244-2)
z = psmcf>' = (aS + h) sin <t>,
where a is the equatorial radius of the spheroid, and C and S are auxiliary functions
that depend on the geodetic latitude and the flattening / of the reference spheroid.
It follows from the properties of the ellipse that
C = (cos2 0 + (1 - ff sin2 <pym, S=(l- ffC. (3.244-3)
Geocentric coordinates may be calculated directly from geodetic coordinates
and conversely. It is easier to use an iterative procedure for the inverse calculation
(see Section 4.2.2). Series expansions, which contain terms up to /3, for 5, C, p, and
4> — 4>' are
3 / GELESTIAL REFERENCE SYSTEMS 1 27
+ (^/2-|/3)cos4^-^/3cos60)
C=l + -f+-f + -f-(-f+-f + -f) cos20
2J 16J 32J \2J V 64J J
+ (J^f + ^/3) cos 40 - cos 60,
\ -(^ + |/3)cos4^ + H/3cos60,
0 _ 4,' = ( / + l-f^j sin 20 - Q/2 + ^/3) sin40 + sin 60. (3.244-4)
The expressions for p and 4> - 4>' are f°r points on the spheroid (h = 0), and the
latter quantity is sometimes known as the "reduction in latitude" or "the angle of
the vertical," and it is of order 10' in midlatitudes. To a first approximation when
h is smaU, the geocentric radius is increased by hi a and the angle of the vertical is
unchanged. The height h refers to a height above the reference spheroid and differs
from the height above mean sea level (i.e., above the geoid) by the "undulation of
the geoid" at the point.
3.25 Aberration
The velocity of light is finite, and so the apparent direction of a moving celestial
object from a moving observer is not the same as the geometric direction of the
object from the observer at the same instant. This displacement of the apparent
position from the geometric position may be attributed in part to the motion of
the object, and in part to the motion of the observer, these motions being referred
to an inertial frame of reference. The former part, independent of the motion of the
observer, may be considered to be a correction for light-time; the latter part, inde-
pendent of the motion or distance of the object, is referred to as stellar aberration,
since for the stars the normal practice is to ignore the correction for light-time. The
sum of the two parts is called planetary aberration, since it is applicable to planets
and other members of the solar system.
3.251 Light-Time In Figure 3.251.1 £ is a stationary observer at time /, and P
is the position of a celestial object also at time f. The dotted curve represents the
orbit of P. The light which is received at E at time / was emitted by the celestial
object when it was at P' at time (f - r), where r is the light-time; i.e., the time
128
EXPLANATORY SUPPLEMENT
£(0 A E V B
Figure 3.251.1 Figure 3.252. 1
Light-time aberration Stellar aberration
taken for the light to travel from P' to E. The direction EP' is called the geometric
direction of the object allowing for light-time.
Light-time is calculated iteratively, with a first approximation r, calculated
from the geometric distance between the object and the observer at time r, and
the next approximation calculated from the distance between P' at (/ — T[ ) and
the observer at time t. No allowance has been made here for the relativistic delay
caused by the Sun's gravitational field. However, the apparent-place algorithm given
in Section 3.31 (3.315) includes this effect, which is of the order of 1 microarcsecond.
3.252 Stellar Aberration In general, the observer at E will be moving with a ve-
locity V. The apparent change in the geometric direction of the celestial object at
P' due to the orbital motion of the Earth about the barycenter is called stellar
aberration. In Figure 3.252.1, p is a unit vector in the geometric direction EP',
i.e., in the direction of the body at time t allowing for light-time (but ignoring the
effect of light deflection). The observer is moving with a velocity V relative to the
stationary frame, and at time t will observe the body at P" in the direction pi,
where P'EB - 9 is the angle between the direction of motion and p in the stationary
frame, and P'EP" - A9 is the displacement due to aberration in the moving frame,
which is always toward the direction of motion.
3 / CELESTIAL REFERENCE SYSTEMS 1 29
The classical Newtonian expression for the direction of the source as seen by
the moving observer is obtained by vector addition of velocities as follows:
Taking the scalar part of the vector cross product of p with Equation 3.252-1, then
(V/c)sin0 V . . 1 fV\2 . _
sinA6>=-^ r = -sm fl--(-j sin20 + -,
v/(l + 2(V/r) + (V/c)2) c
since |p a p, = sin AO, |pAp|=0, and
= — sin (9. (3.252-2)
c
The term of order Vic is about 0.0001 or 20"; the term (Vic)1 has a maximum
value of about O'.'OOl.
In special relativity, the velocity of light is constant in the moving and station-
ary frame, and the Lorentz formula for the addition of velocities applies. Hence
_ 3-'p + (V / c) + (p-V / c)(V / c)/(l + ft-') (3.252-3)
Pl 1+p-V/c
where /?"' = y/l - (Vic)2, again taking the modulus of the vector cross products
of Equation 3.252-3 with p, then
(VI c) sin 6 + \(V I c)2 sin 20 / (1 + p~l)
sin Ad = . 2
1 +(V/c)cos0
= -sm6--(-) sin20+-. (3.252^)
c 4 \ c /
which shows that special-relativistic aberration and classical Newtonian aberration
agree to order Vic (mas precision). However, it is recommended that special-
relativistic aberration (Equation 3.252-3) be used, particularly when high precision
is required (see also apparent- place algorithm, Sections 3.315 and 3.317).
The motion of an observer on the Earth is the resultant of diurnal rotation of
the Earth, the orbital motion of the Earth about the center of mass of the solar
system, and the motion of this center of mass in space. The stellar aberration is
therefore made up of three components, which are referred to as diurnal aberration,
annual aberration, and secular aberration. The stars and the center of mass of the
solar system may each be considered to be in uniform rectilinear motion; in this
case the correction for light-time and the secular aberration are indistinguishable,
and the aberrational displacement due to the relative motion is merely equal to the
proper motion of the star multiplied by the light-time; it is constant for each star,
1 30 EXPLANATORY SUPPLEMENT
and in general, is not known, and is therefore ignored (see Section 3.23 on space
motion for further discussion). The term "stellar aberration" is sometimes loosely
used in contexts where "annual aberration" should strictly be used.
3.253 Classical Annual Aberration In accordance with recommendations of the
International Astronomical Union (1952) the annual aberration is calculated as
from 1960 from the actual motion of the Earth, referred to an inertial frame of
reference and to the center of mass of the solar system. The resulting aberrational
displacement AO may be resolved into corrections to the directional coordinates by
standard methods. If, for example X, Y, and Z are the components of the Earth's
velocity parallel to equatorial rectangular axes, the corrections to right ascension
and declination, referred to the same equator and equinox, in the sense "apparent
place minus mean place" are, to second order in Vic,
X Y
cos 6 Aa = sin q + - cos a
c c
1 . ^ . M*i
+ -^(Xsino. - Kcosq)(Xcos a + Ysin a) sec 6 +
tr
X Y Z
A6 = — cos a sin 6 sin a sin 6 + - cos 6
c c c
1 . • , ' ■
r-(^sina — Y cos a) tan 6
2cl
+ — (Xcqs 8 cos a + YcosS sin a + Zsin 8)
x (Xsin 6 cos a + Ysin 6 sin a — Zcos 6) + ■■■ . (3.253-1 )
These equations are usually used to first order in (Vic), and ignoring the Z term are
expressed in terms of the Besselian day numbers C and D (Section 3.342), which
represent the classical annual aberration terms. An account of how C, D, and the
E-terms of aberration have been calculated in the past is given in Section 3.53.
Atkinson (1972) gives a useful algorithm using a truncated series for calculating C
and D to 0'.'001.
To a lower precision it is possible to use the expressions
X = +0.0172 sin A, Y = -0.0158 cos A, Z = -0. 0068 cos A (3.253-2)
for the barycentric velocity of the Earth with respect to the mean equator and
equinox of J2000.0, where A is the apparent longitude of the Sun.
Also to first order in VI c, we obtain for the aberration in right ascension and
declination due to the unperturbed elliptic component of the orbital motion of
3 / CELESTIAL REFERENCE SYSTEMS
131
the Earth with respect to the Sun, obtained as a correction to be applied to the
geometric place a, 6 in order to obtain the apparent place a(, 6,:
ct\ — a =
6 =
—(k sin © + Ke sin 77) sin a sec 6
— (k cos 0 cos e + Ke cos II cos e) cos a sec 6 (3.253-3)
— (k sin Q + Ke sin II) cos a sin 6
- ( k cos 0 cos e + Ke cos II cos e)(tane cos 6 - sinasin<5),
where © is the true geometric longitude of the Sun, e and 77 are the eccentricity
and longitude of perigee of the solar orbit, e is the mean obliquity of the ecliptic
and k is the constant of aberration (see Equation 3.253-4). The second term in each
factor in Equation 3.253-3 depends explicitly on the eccentricity and represents the
components of the displacement due to the departure of the elliptic orbital motion
from a circle. The component of the aberration that depends on e is known as
elliptic aberration.
The constant of aberration k is the ratio of the mean orbital speed of the Earth
to the speed of light, where perturbations and the motion of the Sun relative to the
barycenter are neglected. It is derived from
K = nal (cv 1 - e2),
(3.253-4)
where c is the speed of light, a is the mean distance of the Earth from the Sun, n
is the mean motion, and e is the eccentricity of the orbit.
The annual aberration due to the barycentric motion of the Sun or to the action
of any particular planet may be obtained, when the ecliptic latitude of the planet
is neglected, from
nma , . . ,
«i — a = (sin q sin / + cos e cos a cos /) sec 6,
c
£ c nma
d\ -o- — (cos a sin 6 sin / + cos /(sin c cos 6 -cose sin a sin <5)), (3.253-5)
where / is the heliocentric ecliptic longitude of the planet, m is the ratio of mass of
the planet to Sun, a is the mean distance, and n is the mean motion. The coefficients
to be used are given in Table 3.253.1. The Sun's aberration in longitude, assuming
unperturbed elliptical motion, can be given as
AA = -ft sec /?(l+e cos v) = na(l - e1) I R,
(3.253-6)
where k is the constant of aberration; a, e, and v are the semi-major axis, eccen-
tricity, and true anomaly of the Earth's orbit; and f3 and R are the Sun's latitude
132
EXPLANATORY SUPPLEMENT
Table 3.253.1
Coefficients (nma / c) for the
Major Planets
Venus
CT0001
Earth
craooi
Jupiter
rroo86
Saturn
o:'ooi9
Uranus
o:'ooo2
Neptune
0:0002
and true radius vector. The e2 term affects only the fifth significant figure; so until
the constant of aberration was specified to five significant figures and the longitude
of the Sun was given to three decimal figures, this term was ignored.
Measurements of radial velocity may be reduced to a common origin at the
barycenter by adding the component of the Earth's velocity in the direction of the
object, that is, by adding
Xcosacos<5 + 7 sin a cos <5 + Zsin 6. (3.253-7)
3.254 Diurnal Aberration The rotation of the Earth on its axis carries the ob-
server toward the east with a velocity up cos 0', where w is the equatorial angular
velocity of the Earth (the standard value of u is given in Section 3.353; if a is the
equatorial radius, then au> = 0. 464 km/s is the equatorial rotational velocity of the
surface of the Earth) , and p and 8X6 the geocentric distance and latitude of the
observer, respectively. The corresponding constant of diurnal aberration is
— - cos 4>' = 0'.'3200 - cos = 0!02133 - cos <f>' . (3.254-1)
c a a a
The aberrational displacement may be resolved into corrections (apparent - mean)
in right ascension and declination:
Aa = 0'02133 - cos <f>' cos h sec 6,
a
AS = 0:'3200 - cos 4>' sin h sin 6, (3.254-2)
a
where h is the hour angle. The effect is small but is of importance in meridian
observations. For a star at transit, h = 0° or 180°, so AS is zero, but
Aq = ±0!02 1 33 - cos <j>' sec S,
a
(3.254-3)
3 / CELESTIAL REFERENCE SYSTEMS
133
where the plus and minus signs are used for upper and lower transits, respectively;
this may be regarded as a correction to the time of transit.
Alternatively, the effect may be computed in rectangular coordinates using
the following expression for the geocentric velocity (r) vector of the observer with
respect to the celestial equatorial reference frame of date:
r =
-up cos 4>' sin(# + A)
up cos <j>' cos(0 + A)
0
(3.254-4)
where 8 is the Greenwich sidereal time (mean or apparent as appropriate), and A
is the longitude (east longitudes are positive).
The geocentric velocity vector of the observer is added to the barycentric veloc-
ity of the Earth's center, to obtain the corresponding barycentric velocity vectors
of the observer (see Section 3.353).
3.255 Planetary Aberration Planetary aberration is the apparent displacement of
the observed position of a celestial body produced by both the motion of the body
and the motion of the Earth. It is often calculated by adding the correction for
stellar aberration (Section 3.252) to the geometric position corrected for light-time
(Section 3.251). On the other hand it may be calculated directly, either by using
the barycentric positions of the body and the Earth at time (? - t), or by using the
barycentric positions and velocities of the body and the Earth at time t as described
below. However, although these latter two methods are simple and widely used, the
first method may be preferred, since it brings out the principle that aberration
depends upon the relative velocity of observer and object.
If we denote the barycentric position of the Earth at time / by EB(r) and the
barycentric position of the planet at time t by uB(t), then the geometric geocentric
position of the planet at time t allowing for light-time is given by
P(r) = uB(f - r) - EB(f).
(3.255-1)
where r = P I c is the light-time and P = |P|. If Pi is the geocentric position of
the planet allowing for planetary aberration, then using classical stellar aberration
(Equation 3.252-1)
P,=P + tEb. (3.255-2)
If we assume that the velocity of the Earth EB and the velocity of the planet uB
are constant during the light-time, then Equation 3.255-1 may be written as
P, = uB(r - r) - EB(f - t),
(3.255-3)
or
134
EXPLANATORY SUPPLEMENT
Pi = Mf) - EB(f) - r(iiB - EB). (3.255-4)
Since Newtonian aberration and special-relativity aberration agree to order Vfc, it
is only when mas precision is required that it becomes necessary to use a rigorous
formula (see Section 3.252).
3.256 Differential Annual Aberration The differential coordinates of a moving ob-
ject with respect to a fixed star will be affected by differential aberration; if Aa,
A6 are the observed differences of the coordinates in the sense moving object minus
star, then the corrections for differential annual aberration are
A(e*i — q) = (Dcosq - C sin a) sec 6Aa + (Dsino + Ccosa)sec<5 tan 6A6,
A(6] - b) = -(Z5 sin a + Ccos a) sin SAa + (D cos a - Csin a) cos 6 AS
- Ctan e0 sin 6 AS, (3.256-1 )
where C and D are the aberration day numbers defined in Section 3.342, and the
units of each term are consistent. The corrections should be applied with those
for differential precession and nutation to give mean positions referred to the same
equator and equinox as those of the stars.
Alternatively the corrections for differential annual aberration to be added to
the observed differences (in the sense moving object minus star) of right ascension
and declination to give true differences are
in right ascension aAa + bA6 in units of 0?001
in declination cAa + dAS in units of O'.'Ol (3.256-2)
where Aa, A6 are the observed differences in units of lm and 1' respectively, and
where a, b, c, d are coefficients defined by
a - -5. 701 cos(H + a) sec 6,
b - -0. 380 sin(// + a) sec 6 tan 6,
c = +8. 552 sin(tf + a) sin <5,
d = -0. 570cos(// + a) cos 6,
fl* = 23.4- (day of year / 15. 2). (3.256-3)
3.257 Differential Planetary Aberration Aberration, because of its dependence on
the relative motions and distances, sometimes has complex effects where two or
more bodies are involved, as, e.g., in eclipses, transits, and the phenomena of satel-
lite systems; and on some past occasions, the determination of these effects has
presented an intricate problem.
3 / CELESTIAL REFERENCE SYSTEMS 1 35
fn a transit of a planet across the disk of the Sun, e.g., the external contacts
occur when the observer is on the conical surface that circumscribes the Sun and
the planet and has its vertex between the Sun and the planet. The internal contact
occur when the observer is on the cone circumscribing the planet and the Sun
having its vertex between the planet and the Earth. The observed contacts are at
the instants when the apparent positions of a point on the limb of the planet and a
point on the limb of the Sun are the same; i.e., the ray of light from the Sun that
reaches the geometric position of the observer at the instant T of contact has grazed
the planet on the way. This ray left the Sun at a previous time T - t2 and reached
the planet at time (T- t2) + ti. The circumscribing cones are formed by the grazing
rays; hence, the points on the Earth and the planet that lie in the same straight
line on one of the cones at the instant of a contact are the geometric position of
the observer at the time T, and the geometric position of the point on the planet at
time T - t2 + r, . Therefore, in the formulas of the theory of transits, for any value
of the time 7", all quantities depending on the time must be derived from the values
of the geometric coordinates (r, /, b) of the planet at T - t2 + T\ and the geometric
coordinates /', b') of the Earth at /.
Similarly, in comparing observed positions of objects in the solar system with
one another or with reference stars, in order to determine the coordinates of a body,
great care is required in correcting the observations for aberration, according to the
means of observation used, and the method of comparison.
In eclipsing binary systems, an apparent variation of the period may be pro-
duced by the variation in light-time with changing distance from the observer due
to an orbital motion of the eclipsing pair with respect to a distant third component
(Irwin, 1959).
3.26 Gravitational light deflection
Gravitational light deflection was predicted by Einstein and first confirmed photo-
graphically at the eclipse of 1919 May by expeditions from Greenwich and Cam-
bridge, England (Dyson et ai, 1920). It has been measured many times, and most
recent high-precision measurements have been made with radio interferometers,
which can observe sources very close to the Sun. Fomalont and Sramek (1975) have
confirmed the deflection predicted by Einstein's theory with an accuracy of 1%,
which is an order of magnitude more accurate than that achieved optically from
eclipse observations. The deflection increases the closer the light path is to the Sun;
at a heliocentric elongation of 90°, however, the deflection has decreased to 0'.'004.
The algorithm for the deflection of light is that of Yallop (1984) as given in The
Astronomical Almanac 1984, which is an adaptation of Murray's (1981) formulas.
The isotropic metric has been assumed. Only the Sun's gravitational field has been
included; each of the planets causes a similar effect that is smaller by a factor
equal to the ratio of the planet's mass to that of the Sun (1/1047 for Jupiter). The
136
EXPLANATORY SUPPLEMENT
S
Figure 3.26.1
Gravitational light deflection
gravitational field of the Earth, also ignored here, can deflect light by a few tenths
of a milliarcsecond for ground-based observers. It may also have to be allowed for
in precise astrometry with Earth satellites.
In Figure 3.26.1, S is the Sun, P the body that is being observed, and E is the
Earth. The unit vectors e and q represent the heliocentric directions of the Earth
and the body, respectively. The heliocentric elongation of the Earth from P is ib
where cos^ = q - e. The geocentric direction to the body P when the light left it
is given by the unit vector p. The dotted arc AEB represents the light-path as it
passes the Earth. The tangent to the light-path at E is XEY. As the light, which
was emitted at P, travels along the path AB, it is always deflected toward the Sun.
At E the direction between p and the tangent to the light-path is A(j> (as shown in
the figure). Einstein's general relativity theory predicts that
. , la sint/j
^ = SET7^>
where E is the distance of the Earth from the Sun, (i is the heliocentric gravitational
constant, and c is the speed of light.
The apparent direction of P is along the tangent to the light path p, , and by
vector addition
(q a e) a p
Pi = p + — — A0, (3.26-2)
sin w
3 / CELESTIAL REFERENCE SYSTEMS 1 37
since (q a e) a p / sin fj> is a unit vector in the plane PES, at right angles to p and
pointing away from the Sun. This equation may be written in the form that is more
useful for computation,
Pi = P + — KP ■ q)e - (e • P)q] ■ (3.26-3)
gi
The dimensionless scalar quantities g\ and gi are
£i = || and g2 = l+q-e, (3.26-4)
where p. = 1.32712438 x lO^mV2, c = 86400 / TA = 173. 144633 Au/day, and
TA = 499.004782 is the light-time for 1 AU in seconds, based on the IAU (1976)
system of constants. The value of g\ is always close to 2 x 10-8, but gi varies
between 0 and +2. The vector pi is a unit vector to order ale1.
Stars represent the asymptotic case when p = q, and Equation 3.26-3 becomes
e - (p e)p
P|=P + £' l+pe °26-5)
When one is applying the light deflection correction in the algorithm for appar-
ent places (see Section 3.316), using the vector EH in the relativistic deflection com-
putation introduces a minor approximation resulting from the use of the barycentric
position of the Sun at the epoch of observation. The resulting error cannot exceed
0. 1 mas in the worst case (object observed at the limb of the Sun with barycen-
tric motion of the Sun orthogonal to the line of sight) and is generally much less.
Furthermore, the deflection algorithm itself results from a first-order development
that assumes small deviations of the photon track from a straight line in Euclidian
space; the error in neglecting second-order effects can reach about 0.5 mas for an
object observed at the Sun's limb (Kammeyer, 1988, private communication).
Figure 3.26.2 shows the magnitude of the deflection of light, as viewed from the
Earth, for planets and stars as a function of the geocentric angular separation of the
observed body from the center of the Sun. Maximum deflection occurs for bodies
that are about to be occulted by the Sun. Minimum deflection for Mercury and
Venus occurs when they are about to transit the Sun. For the other bodies, minimum
deflection occurs at 180° elongation from the Sun. The figure was produced from
Equation 3.26 1 assuming circular orbits.
Table 3.26.1 tabulates the deflection angles A<j) = g\ tan(t/-'/2) (Equation 3.26-1)
for various values of D = 180° - ip (for stars, D approximates the geocentric elonga-
tion). The body disappears behind the Sun when D is less than the limiting grazing
ralue of about 0?25.
138
EXPLANATORY SUPPLEMENT
Of 27
45° 90° 180°
Of 1 1° 10° 100°
Geocentric elongation
Figure 3.26.2
Light from the planets and stars deflected by the sun
The table was produced assuming that the Earth-Sun distance E = 1, and
thus g\ = 01'00407. The variation of the Earth-Sun distance modulates the value
°f gi by less than 2%. This simple form of the gravitational deflection formula
has been noted previously by Shapiro (1967) and by Fukushima (1982, private
communication). In this form, the deflection is not explicitly dependent on the
distance of the emitting body from the Sun or the Earth. Therefore, to an observer
on the Earth, the apparent gravitational deflection is the same for all objects that
lie anywhere on a given line that extends radially outward from the Sun. This result
holds regardless of the orientation of the line with respect to the Earth.
The equations for light deflection have an indeterminacy for light paths starting
beyond the Sun on the extension of the Sun-Earth line; bodies there are hidden by
the Sun's disk and unobservable in any event. For these bodies or the Sun itself,
the deflection can be considered to be zero.
Table 3.26.1
Apparent Deflection Angles
D 0?25 0?5 1° 2° 5° 10° 20° 50° 90°
A<t> 1V866 o:'933 0!'466 01'233 o:'093 0"047 0'.'023 0"009 0'.'004
3 / CELESTIAL REFERENCE SYSTEMS 1 39
3.261 Approximate Light Deflection for Stars The increments to be added to the
calculated right ascension and declination of the star may be evaluated approxi-
mately from
cos D = sin 8 sin Ss + cos 6 cos <5S cos (a — as),
Aa = vmo2ncos6>sin{a-ai
(1 — cosZ))cos<5
A6 = 0:'00407 sin6coS4cos(a-as)-cos^in4 2f)
1 - cos D '
where a, <5, ctS) <5S are the geocentric right ascensions and decimations of the star
and the Sun, respectively. For stars, D, the geocentric elongation, approximates
1 80° — ip, where ip is the heliocentric elongation. These corrections may be included
with the day-number reduction for stars given in Section 3.341.
3.27 Polar Motion
The rotation of the Earth is represented by a diurnal rotation around a reference
axis whose motion with respect to the inertial reference frame is represented by the
theories of precession and nutation. The reference axis does not coincide with the
axis of figure (maximum moment of inertia) of the Earth, but moves slowly (in a
terrestrial reference frame) in a quasi- circular path around it. The reference axis
is the Celestial Ephemeris Pole (normal to the true equator), and its motion with
respect to the terrestrial reference frame is known as polar motion. The Celestial
Ephemeris Pole is the axis about which the diurnal rotation of the Earth is applied
in the transformation between celestial and terrestrial frames, and so must not be
confused with the instantaneous axis of rotation. The maximum amplitude of the
polar motion is typically about 0"3 (corresponding to a displacement of about 9 m
on the surface of the Earth) and the principal periods are about 365 and 428 days.
The motion is affected by unpredictable geophysical forces, and is determined from
observations of stars, radio sources, the Moon, and appropriate Earth satellites,
using relevant techniques including VLBI and laser ranging.
The pole and zero (Greenwich) meridian of the terrestrial reference frame are
defined implicitly by the adoption of a set of coordinates for the instruments that are
used to determine UT and polar motion from astronomical observations. (The pole
of this system is known as the Conventional International Origin.) The position of
this terrestrial reference frame with respect to the true equator and equinox of date
is defined by successive rotations through two small angles x, y, and the Greenwich
apparent sidereal time 6. The angles x and y correspond to the coordinates of the
Celestial Ephemeris Pole with respect to the terrestrial pole measured along the
meridians at longitude 0° and 270° (90° west). Current values are published by the
International Earth Rotation Service; values from 1970 January 1 onward are given
1 40 EXPLANATORY SUPPLEMENT
in section K of The Astronomical Almanac. An 80-year-long series of values on a
consistent basis has been published by the International Polar Motion Service. The
coordinates x and y are usually measured in seconds of arc.
Polar motion causes variations in the zenith distance and azimuth of the Ce-
lestial Ephemeris Pole and hence in the values of the terrestrial latitude (0) and
longitude (A) that are determined from direct astronomical observations of latitude
and time.
The rigorous transformation of a vector ro with respect to the frame of the
true equator and equinox of date to the corresponding vector r with respect to the
terrestrial frame is given by the formula
r = R2(-x)R](-y)R3(e)r0; (3.27-1)
conversely,
r0 = R3(-0) R,(y) R2(*) r, (3.27-2)
where Equation 3.27-1 represents a rotation of 6 about the z-axis, followed by a
rotation of — y about the x-axis, and finally a rotation of — x about the y-axis (see
Section 11.4). The vector r could represent, for example, the coordinates of a point
on the Earth's surface or of a satellite in orbit around the Earth.
Alternatively the transformation can be expressed as a variation of the longi-
tude and latitude of a point, and to first order, the departures from the mean values
Am, <j>m (corresponding to vector Tq) are given by
A<f> = xcos Ara — y sin Am and AA = (jrsin Am +ycosAm)tan. <j>m. (3.27-3)
The variation in longitude must be taken into account in the determination of
Greenwich mean sidereal time, and hence of UT, from observations.
3.28 Refraction
Atmospheric refraction is included in the The Astronomical Almanac in only a few
topocentric phenomena, such as the times of rising and setting of the Sun and
Moon, and in theory in the predictions of local circumstances of eclipses. How-
ever, for observational reductions the effect of refraction must be included. In this
section we give a low-precision formula, valid at the horizon, that is particularly
useful for navigation; a formula valid for zenith distances less than 70°, which takes
into account the amount of water vapor in the atmosphere; and an algorithm for
evaluating the integral in the classical formulation using quadrature. This latter
algorithm, which allows for the principal change of refraction due to variations of
3 / CELESTIAL REFERENCE SYSTEMS 1 41
temperature, pressure, and humidity and even temperature lapse rate, gives a well-
defined reference value. Thus, for example, residuals of the observed values from
the reference values could be used to model the local variation of refraction.
3.281 Refraction — Numerical Integration In this section we give a precise algo-
rithm for calculating refraction using numerical quadrature. The real accuracy is
very dependent on the conditions, particularly for large zenith distances, and on
the atmospheric model.
The atmosphere is assumed to be spherically symmetric and in hydrostatic
equilibrium, and to obey the perfect gas law for the combined mixture of dry air
and water vapor, and also for the dry air and water vapor separately. The two
layers of the atmosphere are the troposphere and the stratosphere. The troposphere
extends from the surface of the Earth to the tropopause, which is assumed to be
11 km (hi) from the Earth's surface. In this region the temperature decreases at a
constant rate (a = 0. 0065° km^ ) and the relative humidity is constant and equal
to its value at the observer. In the stratosphere the temperature remains constant
and equal to the temperature at the tropopause (T,) and there is no pressure due
to water vapor. The upper height (hs) of the stratosphere is taken as 80 km, since
above this height refraction becomes negligible.
The total bending of a ray is given by
I dz_ (3.281-1)
Jo n + ran I dr
This integral is a transformation of the usual refraction integral. It has been rec-
ommended by Auer and Standish (1979) because it is more suitable for numerical
quadrature; it is a more slowly varying function over the whole range of z and re-
moves the problem at z = 90°. However, there is a discontinuity in the function
dnl dr at the tropopause, and so the integral is evaluated in two parts, from z = Zs
to Zis in the stratosphere, and from z = Zt to Zo in the troposphere.
The following procedure gives the steps required to calculate the total refrac-
tion, £, i.e., the sum of the refraction caused by the troposphere (£,) and the strato-
sphere (&), from
P f dz+ P fdz (3,281-2)
J JO -' Zi,
to a specified tolerance e, where / is the integrand given in Equation 3.281-1.
1 42 EXPLANATORY SUPPLEMENT
Step ii The parameters of the Earth and the atmospheric model, which are as-
sumed to be constant are
R = 8314.36 A/,, = 28.966 MK = 18.016 6 = 18.36
rt = 6378 1 20 ft, = 1] 000 hs = 80000 a = 0. 0065 (3.28 1 -3)
Step 2: The parameters of observation— the initial conditions. The observed zenith
distance, zo, of an object with wavelength of A/zm is recorded by an observer at
latitude 0, height h0 meters above the geoid i.e., r0 = re + h0 meters from the cen-
ter of the Earth. The meteorological conditions at the time of observation are the
temperature 7^ K, the pressure P0mb and the relative humidity Rh.
Step 3: Calculate the parameters dependent on the initial conditions and the
model of the atmosphere. Those denoted by Q are needed at each step of the
integration.
a fwrtnA h62%% 0.0136\
A = 287. 604 + ; — +
V X2 X4 J
x 10"6.
/Vi = /?,,( 7b / 247. 1)*,
g = 9. 784( 1 - 0. 0026 cos 2d - 0. 0000 0028 h0),
1.6288 0.0\36\ 273. 15
1013.25
C, = a, C2=gMd/ R, C3 = C2 / C, = 7,
C4 = 6, C5= />,,<>( I -Mai A/H,)7 / (6 - 7),
C6 = A(P0 + Cs) / To, C7 m (ACs + 11. 2684 x 1 0~bPM) I T0,
Cs = a(7 - 1 )C6 / To, C9 = a(6 - 1)C7 / 7b. (3.281^1)
Step 4: Use the following expressions to calculate r, n, and dn i d>\ which depend
on the atmosphere, and thus evaluate the integrand / at each step in z along the
path of the ray.
The value of r corresponding to the current step in zenith distance z is found
by solving Snell's law nrsinz = nQr0 sinz0 using Newton- Raphson iteration thus:
n+i = ti -
n,n - nQr0 s'mzu I sin;
n, + ndtij I dr,
for i = 1,2,..., (3.281-5)
where rx is the value of r calculated at the previous step of the integration. Con-
vergence is rapid, so the initial estimate r\ is not critical. Four iterations should be
sufficient.
3 / CELESTIAL REFERENCE SYSTEMS
143
In the troposphere at distance r < r, from the center of the Earth, the temper-
ature T, the refractive index n, and dn! dr are calculated from
T = T0 - a(r - r0),
n=\ +
dn
~dr
L
f:,
1-2
■ ^
T0
6-2
r
n'
(3.281-6)
In the stratosphere at a distance r where r, < r < rs and the temperature is given
by T, = T0 - a(r, - ro) the refractive index n and dn/ dr are calculated from
n = 1 + (/t, - 1 ) exp ( - C2
dn _ O
dr ~
r- r,
T,
r —
T (n, - l)exp^ - C2- ?
The integrand is
r'n'fr
rdn I dr
n + rdn I dr
(3.281-7)
(3.281-8)
Step 5: Calculate the following parameters which are required for the limits of the
two integrals. At the observer, the observed zenith distance is zo\ set r0 = re +h, and
calculate «0, dno/dr0, fo- At the tropopause, in the troposphere calculate r, - re +h,,
n,, dn, I dr,, /,, and using Snell's law
■ nnro sinzn
Zt = sin"1 — K (3.281-9)
n,r,
At the tropopause, in the stratosphere r,x = r,, and calculate nu, dnls I dr,s, /„, and
Zij. At the limit of the stratosphere calculate rs-re+ hs, ns, dns I drs, fs, and zs-
Step 6: Integrate the function over the required interval using numerical quadra-
ture, e.g., Simpson's rule, by forming the summations over equal steps of z. In the
troposphere S = (z, - z0)/ 16 is a convenient step length to start with. Repeat the
integration, halving the step size each time, until there is no significant difference
between two consecutive values and the tolerance permitted (i.e., 0.5e). Repeat this
step for the stratosphere. Thus the total amount of refraction will be £ = £, +
3.282 Saastamoinen's Refraction Formula This formula, valid for zenith distances
down to 70°, was devised by Saastamoinen (1972), and it is equivalent to the re-
fraction tables in The Star Almanac. Given the observed zenith distance, zo, the
-) 44 EXPLANATORY SUPPLEMENT
temperature, pressure, and partial pressure of water vapor, (7o, Pq, Pwo), the refrac-
tion for a wavelength of 0.574/im, i.e., visible light, and for an observer at sea
level is
f - 1 6"27 1 Q tan za { 1 + 0. 0000 394 Q tan2 zo)
- 0"0000 749 P0(tm zq + tan3 zo),
where Q - (Pq - 0. 156fU) / % (3.282-1)
A formula to calculate Phq in terms of the relative humidity is given at the start of
Step 3 in Section 3.281. Saastamoinen gives correction tables for other wavelengths
and heights above sea level.
3.283 Low-precision Refraction Topocentric phenomena such as the times of ris-
ing and setting, and navigational reductions require the amount of refraction ac-
curate to 0'.5. The following formula is dependent on the apparent altitude, H;
i.e., the observed altitude corrected for the index error of the sextant and dip (see
Section 9.331), and on the temperature and pressure, and gives the amount of
refraction, /?, in degrees;
R = (±2*L\ 0?0167
\T + 2n) tan(// + 7.31/(// + 4.4))'
where T,P are the temperature in degrees Celsius and the pressure in millibars.
If T and P are unknown, assume the term in the first bracket is unity. The effect
of refraction is removed from an observation by subtracting R from the observed
altitude. When only the true altitude is known, it is possible to iterate using the
formula above, in order to find the observed altitude. This is necessary, particularly
at low altitudes, when the refraction correction is large.
Alternatively, at low altitudes, when H < 15°, the following may be used:
/ P \ 0? 1594 + 0. 0196// + 0. 00002//2 283-2)
" \T + TTi ) 1 +0.505// + 0.0845//2
For rising and setting phenomena a constant of 34' is used for the horizontal
refraction; thus for the center of the Sun or Moon to be coincident with the horizon
the true altitude will be —34'.
The observed right ascension and declination, (a, 6), of an object may be
corrected, to first order, to give the true position Q\,6i, using the following formulas:
Q] = Q
RsecS sinC
6i = 6 — R cos C,
(3.283-3)
3 / CELESTIAL REFERENCE SYSTEMS 1 45
where R is the refraction and C is the parallactic angle— i.e., the angle at the object
in the spherical triangle pole-object-zenith. Thus when the object is on the meridian
C = 0°, and the amount of refraction, R, is a direct correction to declination only,
but the right ascension is unaffected.
Refraction tables based on a standard temperature and pressure are printed in
various publications, for example, Abalakin (1985), Pulkova (1956), together with
correction tables for other temperatures and pressures. The standard conditions
used in The Nautical Almanac are T = 10°C and P = 1010 mb; in The Star Almanac
T=1°C and the pressure P = 1005 mb.
3.3 APPARENT AND TOPOCENTRIC PLACE ALGORITHMS
This section presents a set of algorithms for calculating apparent and topocentric
places of planets and stars. The methods are given in a stepwise form using matrix
and vector operations which can then be programmed using a set of subroutines,
each of which handles one particular aspect of the reduction. Full details on the
various topics are discussed in the appropriate part of Section 3.2. The algorithm
for calculating apparent places of planets and solar-system bodies is given first,
because it is the more comprehensive. The procedures for calculating apparent
places (Sections 3.31 and 3.32), topocentric places (Section 3.35), and differential
astrometry (Section 3.4) are based on Kaplan et al. (1989).
The following notation is used in the algorithms. The subscript B refers to the
solar-system barvcenter, u = |u| means calculate the square root of the sum of the
squares of the components, and (u • r) indicates the scalar product, i.e., the sum of
the products of their corresponding components.
3.31 Apparent-Place Algorithm for Planets
The algorithm used to compute the apparent place of a planet or other solar-
system bodv at an epoch of observation f', given its ephemeris with origin at the
solar-system barycenter and its coordinates referred to the Earth's mean equator
and equinox of a reference epoch, to, can be succinctly represented as
u4(f') - N(0P(0/lg [uB(r - t) - EB(f)]], (3.31-1)
where
t> is the epoch of observation, in the TDT timescale;
i is the epoch of observation, in the TDB timescale;
to is the reference epoch to which the ephemeris is referred, e.g., J 2000.0;
r is the light travel time from the planet to the Earth, in the TDB timescale,
for light arriving at the epoch of observation f,
1 46 EXPLANATORY SUPPLEMENT
uB(/ — t) is the barycentric position of the planet at epoch t - r, referred to the
mean equator and equinox of /oi
EB(t) is the barycentric position of the Earth at the epoch of observation /.
referred to the mean equator and equinox of to;
g[...] is the function representing the gravitational deflection of light;
/[...] is the function representing the aberration of light;
P(f) is the precession matrix, a rotation from the mean equinox and equator
of /o to the epoch of observation t\
N(r) is the nutation matrix, a rotation from the mean equinox and equator
of date to the true equinox and equator of date t;
u4(/') is the apparent geocentric place of the planet at the epoch of observation
t' represented as a three-dimensional position vector (x,y, z) and referred
to the true equinox and equator of observation.
This expression is schematic; the full functional forms of / and g, the elements
of the P and N matrices, and other auxiliary calculations are not indicated.
3.311 Relevant Time Arguments
Step a: Express the epoch of observation t1 as a TDT Julian date.
Step b: Compute T', the number of Julian centuries in the TDT timescale from
J2000.0 TDT (JD 2451545.0) TDT.
T' = (t' - 245 1545.0)/ 36525 (3.311-1)
Step c: Compute the mean anomaly m of the Earth in its orbit, in radians, at the
epoch of observation
m = (357. 528 + 35999. 050 T) x 2tt 1 360. (3.31 1-2)
Step d: Compute s, the difference, in seconds, between the clock reading in the two
timescales (in the sense TDB -TDT), and t, the TDB Julian date corresponding
to the epoch of observation
5 = 0. 001658 sin(w + 0.0167l sinw),
/ = /' + s I 86400.
(3.311-3)
3 / CELESTIAL REFERENCE SYSTEMS 1 47
Step e: Compute 7*, the number of Julian centuries in the TDB timescale elapsed
since J2000.0 TDB (JD 2451545.0 TDB)
T={t- 2451545.0)/ 36525. (3.311-4)
In the expression for s, the lunar and planetary terms of order 10-5 seconds have
been ignored. Furthermore, the expression for m (Step c) strictly requires a time
argument in the TDB, not the TDT, timescale. See Moyer (1981) for a complete
discussion.
However, the algorithm given above is much more precise than is required for
the computation of apparent places of stars and most solar-system bodies. For
stellar apparent places, set s = 0, / = t1, and T =T' with negligible error. For solar-
system bodies, the same approximation can be used for all bodies except the Moon
and close-approaching comets and asteroids, where the error in using the t = t'
approximation may approach O'.'OOl in very unfavorable circumstances.
3.312 Ephemeris Data for the Earth and Sun
Step f: Extract from the ephemeris, for the time t, the barycentric position and
velocity of the Earth, EB(f) and EB(r), and the barycentric position of the Sun,
Sb(0j referred to the Earth's mean equator and equinox of the reference epoch ?o-
Also form the heliocentric position of the Earth EH(f) = EB(r) - SbU)- Note that the
barycentric position of the Earth is that of the center of mass of the Earth, and
not that of the Earth Moon barycenter. All position vectors are in AU and velocity
vectors in AL'/day.
The standard ephemeris of the major bodies in the solar system is the Jet
Propulsion Laboratory ephemeris designated DE200 (Standish 1982a). The posi-
tions of the Sun, Moon, and planets given in The Astronomical Almanac and other
international almanacs are now obtained from this ephemeris. Values of the com-
ponents of the vectors EB. EB, and —Eh at 1-day intervals are tabulated in The
Astronomical Almanac. A set of analytical planetary theories fitted to DE200 has
been developed by Bretagnon (1982).
The barycentric position of the Earth is used to form the geocentric position
of the body (see Sections 3.314 and 3.315); the barycentric velocity of the Earth is
used in the aberration computation (Section 3.317); and the heliocentric position
of the Earth is used in the computation of the relativistic gravitational deflection
of light (Section 3.316).
148
3.313 Ephemeris Data for the Planet
EXPLANATORY SUPPLEMENT
Step g: Extract from the ephemeris, for the time /, the barycentric position of the
planet, uB(/), referred to the Earth's mean equator and equinox of the reference
epoch J2000.0.
3.314 Geometric Distance Between Earth and Planet
Step h: Compute d, the geometric distance between the positions of the center of
mass of the planet and the Earth at time t, in AU, from
d = |uB(f) - EB(0|- (3.314-1)
The geometric distance is the quantity tabulated in The Astronomical Almanac as
the "true distance" of solar-system bodies.
Using d compute r, a first approximation to the light-travel time between the
planet and the Earth, as
T = d/c, (3.314-2)
where c is the speed of light expressed in AU/day; its precise value may be computed
from 86400/ ta, where ta is the light-time for unit distance (1 au) in seconds. In the
IAU (1976) system c= 86400/499.004782 - 173. 144633 Au/day.
3.315 Geocentric Position of Planet, Accounting for Light-Time
Step i: Extract from the ephemeris, for the time t - t, the barycentric position of
the planet and the Sun, uB(f - r) and SB(f - r), respectively.
Step j: Calculate U and Q, approximations to the geocentric and heliocentric po-
sition of the center of mass of the planet, respectively, at the epoch of observation
t, from
U = uB(> - r) - EB(0, (3.315-1)
Q = uB(/ - T) - SB(/ - t). (3.315-2)
Step k: Next compute t', a better approximation to the light-travel time between
the planet and the Earth, as
3 / CELESTIAL REFERENCE SYSTEMS
149
Cr' = (/ + (2A//c2)ln|^|, (3.315-3)
where U = |U|, E = |EH|. Q - |Q|, H is the heliocentric gravitational constant and
c is the speed of light in AU/day (Equation 3.314-2). Using the IAU 1976 system,
fi= 1. 32712438 x 1020m3s-2.
Compare r' with t; if they are identical within some small tolerance, continue
to Step 1. If they are not, then replace the value of r with the value of r' and repeat
Steps i through k until the light-time converges to within the tolerance permitted.
Since the speed of bodies in the solar system is small compared to the speed of
light, this process converges rapidly.
The tolerance permitted depends on the precision desired in the final coordi-
nates and the apparent angular speed of the body as viewed from Earth. The most
rapidly moving objects in the sky are the Moon (angular rate = 0'.'5 per second),
Mercury (angular rate at inferior conjunction 0!'05per second), and the Sun (an-
gular rate 0'.'04per second). However, occasionally an Earth-crossing asteroid or
comet may exceed these rates for short periods of time. For a computational preci-
sion of one milliarcsecond, therefore, the light-time convergence tolerance must be
0?002 = 2 x 10 8 days or less; we suggest 1 x 10~8 days.
Kaplan et al. (1989) ignore the second term in Equation 3.315-3, (order 1 micro-
arcsecond), which is due to the relativistic delay caused by the Sun's gravitational
field.
Step 1: Once r has converged, set U = uB(r-r)-EB(f) and Q = uB(f-T)-SB(/-r),
the geocentric and heliocentric positions of the body at time t — r.
3.316 Relativistic Deflection of Light
Step m: Form the following unit vectors and dimensionless scalar: quantities
u = U/|U|, q = Q/|Q|, e = EH/|EH|,
2 (3.316-1)
where the constants fi and c are given in Steps k and h. A detailed explanation on
light deflection is given in Section 3.26.
"150 EXPLANATORY SUPPLEMENT
Step n: The deflected position of the body relative to the geocentric inertia] frame
that is instantaneously stationary in the space-time reference frame of the solar
system (the natural frame) is then
U| = 'u' (u + f! [(u ■ q)e _ (e ■ u) qI) • (3'3 1 6~2)
The vector u / )U| is a unit vector to order pi c2.
For the Sun or bodies lying behind the Sun on the Sun-Earth line, the light deflec-
tion can be considered zero, so that u = U.
3.317 Aberration of Light
Step o: Form the following quantities:
p = ttl/|»i|, V = EB(r)/c,
0-l*Vl-V*, /,=(p-V), /2 = i+/,/(i+/3-i); (3.317-1)
where the velocity (V) is expressed in units of the velocity of light and is equal to
the Earth's velocity in the barycentric frame to order V2.
Step p: The aberrated position of the body in the geocentric inertial frame that is
moving with instantaneous velocity (V) of the Earth relative to the natural frame
is then given by the vector
»2 = 0?~Vi +/2|ui|V)/(l +/,). (3.317-2)
The above algorithm includes relativistic terms (Murray. 1981) (see Section 3.252).
which are of the order 1 milliarcsecond. Therefore, for many applications one may
use the much simpler classical formula
u2 = Ui+|u,|V. (3.317-3)
3.318 Precession
Step q: Evaluate the three fundamental precession angles zA, and 0A using
the equations given in Section 3.211 with epoch eF = J2000.0 and eD = t and the
associated matrix P (Equation 3.21-8).
3 / CELESTIAL REFERENCE SYSTEMS 151
Step r: Transform the coordinates from the fixed reference epoch to - J2000. 0 to
the epoch of observation t, by applying the precession matrix P to the vector u2:
u3=Pu2. (3.318-1)
3.319 Nutation
Step s: Evaluate the mean obliquity of the ecliptic, e$, the two fundamental nuta-
tion angles, Aw, Ac and the true obliquity of the ecliptic e = €q + Ae, for the epoch
of observation t, and calculate the associated nutation matrix N. Full details are
given in Section 3.222.
Step t: Transform the coordinate system to that defined by the Earth's true equa-
tor and equinox at the epoch of observation /, by applying the nutation matrix N
to vector 113 :
u4 = Nu3, (3,319-1)
where 114 is the apparent geocentric position vector of the planet at the epoch of
observation t' .
The combined precession-nutation matrices given in The Astronomical Almanac
(R = NP) may be used in place of Steps q through t above; thus 114 = Rui, where
the values of the elements of R must be interpolated to the epoch of observation.
3.31 10 The Apparent Position in Spherical Coordinates
Step u: Compute the object's apparent geocentric right ascension, a, and decli-
nation, S, at the epoch of observation f, using the three components of the vector
U4, i.e., x, y, and z:
a = tan-'^ and 6 = tan"1 —=J=. (3.3110-1)
* \/x2+y1
Note: U4 is not a unit vector. Most computers have a double argument function
for inverse tangent (e.g., ATAN2) which will provide the correct quadrant if the
numerator and denominator are entered as separated arguments.
152
3.32 Apparent-Place Algorithm for Stars
EXPLANATORY SUPPLEMENT
The algorithm used to compute the apparent place of a star at an epoch of ob-
servation f', given its mean place, proper motion, and other data (as available) at
reference epoch to, can be succinctly represented as
04(0 = N(0 P(r)/[g[uB(r0) + (t - /0)uB('o) - E„(r)]], (3.32-1 )
where
t' is the epoch of observation, in the TDT timescale;
/ is the epoch of observation, in the TDB timescale;
/0 is the reference epoch and equinox, e.g., .12000.0, of the star catalog, in
the TDB timescale;
uB(f0) is the catalog mean place of the star at the reference epoch to, represented
as a three-dimensional position vector in AU, with origin, solar-system
barycenter, and coordinates referred to the mean equator and equinox
of t0;
Ub('o) 's the space motion vector (in Au/day) of the star at the reference epoch
to, obtained from the catalog proper motions, parallax, and radial veloc-
ity;
Eb is the barycentric position of the Earth at the epoch of observation t.
referred to the mean equator and equinox of to;
g[...] is the function representing the gravitational deflection of light;
/[...] is the function representing the aberration of light;
P(f) is the precession matrix, a rotation from the mean equinox and equator
of to to the epoch of observation f;
N(f) is the nutation matrix, a rotation from the mean equinox and equator
of date to the true equinox and equator of date t;
U4(r') is the apparent geocentric place of the star at the epoch of observation
t' represented as a three-dimensional position vector (x,y, z) and referred
to the true equinox and equator of observation.
This expression is similar to Equation 3.31-1 for planets. Most of the algorithm
is identical; the difference is due only to the more complex motion of a planet
compared to that of a star. In Section 3.31 we had to obtain the position of the
planet from an ephemeris and deal with the light-time problem explicitly; in this
section we will assume uniform rectilinear motion for the star and neglect variations
in light-time as the star moves.
3.321 Relevant Time Arguments
Step 1: Follow Steps a through e in Section 3.311. This gives the value of t, the
TDB Julian date corresponding to the epoch of observation, and T the number of
3 / CELESTIAL REFERENCE SYSTEMS 1 53
Julian centuries in the TDB timescale elapsed from the reference epoch J2000.0
TDB. For stars, skip Step c and simply set t - t'.
3.322 Ephemeris Data for the Earth and Sun
Step 2: Follow Step f in Section 3.312 and extract from an ephemeris, for time /,
Eb(0i Eb(0, and Sb(0) the barycentric position and velocity of the Earth, and the
barycentric position of the Sun. Also, form EH(f) = EB(r) - SB(r), the heliocentric
position of the Earth. All position vectors are in au and velocity vectors in Au/day.
For apparent places of stars accurate to a few milliarcsecond it is necessary
only to obtain the Earth and Sun's position components to three significant dig-
its and the Earth's velocity components to five significant digits. It is therefore
feasible to construct relatively compact closed-form algorithms which provide the
required data (Stumpff, 1980b). The U.S. Naval Observatory has developed com-
puter subroutines, based on a truncated and modified form of Newcomb's theory,
which evaluates the barycentric position and velocity of the Earth, without using
external files. However, for the highest precision, or when the apparent places of
planets are being computed more complex algorithms or external files are required.
3.323 Star's Position and Space Motion Vectors at the Catalog Epoch
Step 3: Given are a and 6, the catalog mean barycentric right ascension and decli-
nation of the star at t0 = 245 1545. 0, the reference epoch J2000.0, and f.ta and fi6 the
corresponding proper motion components in seconds of time and arc, respectively,
per Julian century (of TDB). Also given are p, the parallax of the star in seconds
of arc, and r, its radial velocity in km/s.
Form the barycentric position vector, uB(fo) (in Au), and space motion vector,
"bCo) (AU/day), of the star at the catalog epoch f0, referred to the Earth's mean
equator and equinox of the catalog epoch using Equations 3.23-1 and 3.23-3 in
Section 3.23.
3.324 Star's Position Vector at the Epoch of Observation
Step 4: Compute the barycentric position of the star, at the epoch of observation,
by adding the distance moved during the interval of time between the catalog epoch
and the epoch of observation, as
uB(f) = ub('o) + (*'- 'u)uB('a).
(3.324-1)
154
EXPLANATORY SUPPLEMENT
The assumption is made that there is no explicit correction for light-time in stellar
apparent-place computations (see Section 3.23).
3.325 Geocentric Position of the Star
Step 5: Form the vector U, which represents the geocentric position of the star at
the epoch of observation, t.
U = uB(f)-EB(f). (3.325-1 j
This step introduces annual parallax, and is a shift of the origin from the solar
system barycenter to the Earth's center of mass.
3.326 Relativistic Deflection of Light
Step 6: Follow Steps m through n in Section 3.316, with Q = Ub(0 — SB(f) the
heliocentric position of the star, to obtain the geocentric position of the star Uj,
corrected for relativistic light deflection.
3.327 Aberration, Precession and Nutation
Step 7: Follow Steps o and p in Section 3.317.
Step 8: Follow Steps q and r in Section 3.318.
Step 9: Follow Steps s and t in Section 3.319.
3.328 Star's Apparent Position in Spherical Coordinates
Step 10: Follow Step u in Section 3.3110.
3.33 The Computer Implementation of Apparent-Place Algorithms
The development of the apparent-place algorithms using matrix/vector notation is
rigorous, and allows the planet and star algorithms to use those parts which are
common to both. It also allows users to tailor their program to their own needs, as
well as the testing of alternative, or simplified, algorithms for special purposes.
In this unified approach for planets and stars, position vectors have been used
throughout. Many other algorithms use unit vectors (see The Astronomical Alma-
nac). There are penalties either way, and if the procedures are carefully implemented
the results will be identical.
The method given here involves a series of steps. However, if the observation
time remains the same, quantities such as the time arguments, the ephemeris data
3 / CELESTIAL REFERENCE SYSTEMS 1 55
for the Earth, and the precession and nutation matrices can be saved and re-used.
For example, when there are a large number of apparent places to be computed,
it is more efficient to compute the places of all bodies at a given observing epoch
before moving to a new observing epoch. The major computational burden in the
algorithm involves the retrieval of data from the planetary ephemeris in Steps f
through 1 and the evaluation of the two nutation angles A^ and Ae in Step s.
If computational time becomes critical (such as within telescope-control sys-
tems or in microcomputer implementations), the method of obtaining the ephem-
eris and nutation data must be considered. Self-contained algorithms can often be
simplified by truncating small terms from the series, if high precision in the final
apparent place is not required. Consideration should also be given to precomputing
the required data for fixed intervals and storing the data in an external file that can
be efficiently accessed and interpolated. Planetary ephemeris data are frequently
distributed in this form anyway. Precomputing and storing the elements of the com-
bined precession-nutation matrix (see end of Step t) is also feasible; see Section B of
The Astronomical Almanac. The convenience of self-contained algorithms must be
weighed against the number of calculations required, the accuracy of the algorithm,
and the accuracy needed for the final result.
3.34 Apparent-Places-Day-Number Technique
The 1976 IAU resolutions on the astronomical reference frame included two state-
ments about the procedures for computing apparent places and the reduction of
observations. First, reductions to an apparent place shall be computed rigorously
and directly without the intermediary of the mean place for the beginning of the
year whenever high precision is required. Second, stellar aberration shall be com-
puted for the total velocity of the Earth referred to the barycenter of the solar
system and the mean places shall not contain E-terms. Implicit in that recommen-
dation is the fact that, when accurate results are desired, day numbers should not
be used. Thus, this section on day numbers is included as a means to help unravel
what may have been done to observations in the past and to understand the proce-
dures that were followed in the past for historical purposes, and for those who can
be satisfied with a reduced level of accuracy, e.g., for making rapid calculations of
star positions in real time for pointing telescopes.
Before we explain the methods of day numbers, an additional warning concern-
ing the past practice for computing day numbers seems advisable. Specifically there
have been a number of changes in the method used for computing the aberration
day numbers, C and D. A detailed discussion is given in Section 3.53.
The day-number technique may also be used for objects within the solar system,
but the geocentric position of the body is required at the time t — At, where At is
the light-time. Thus the proper motion, and annual parallax corrections must be
omitted.
1 56 EXPLANATORY SUPPLEMENT
The calculation of the apparent place of a star for a date t+T, where t represents
a fixed epoch such as the beginning of the Besselian year or, as is current practice,
the middle of the Julian year, and r represents a fraction of the year, first requires
the calculation of the mean place for the mean equinox and epoch of t. The reduction
then involves the application of corrections for precession, from the epoch t for the
interval f, nutation, stellar aberration, annual parallax, proper motion, and orbital
motion.
Proper motion, orbital motion, and stellar aberration do not affect the frame
of reference, but cause changes in the actual direction in which the star is observed;
the corresponding corrections must therefore be calculated with respect to a par-
ticular reference system and applied to the position of the star in the same system.
Precession and nutation, however, are changes in the frame of reference and do not
affect the actual direction in which the star is observed. These two corrections, and
that for annual aberration, are sufficiently large to make their order of application
of significance if cross-product terms are neglected. The corrections for parallax,
proper motion, and orbital motion are generally very small and can be applied at
any convenient stage.
Since nutation is calculated from the longitudes of the Sun and Moon referred to
the mean equinox of date, it is (theoretically) necessary to apply precession before
nutation. There are then two methods for calculating the reduction to apparent
place; if precession and nutation are applied first {Method 1), then the aberration
correction should strictly be applied to a fixed star whose coordinates (referred to
the moving frame of reference) are continuously changing. If aberration is applied
first (Method 2) then the corrections for precession and nutation should strictly be
applied to the changing position of the star.
The two methods give identical results, of course, but for systematic calcula-
tion one must apply corrections for precession, nutation, and aberration to a fixed
star, any residual corrections (if appreciable) being applied separately. As might be
expected, the largest correcting term is the same for both methods, but the other
terms differ. An analysis of the magnitude of the residual terms, taken in conjunc-
tion with the second-order terms of precession, nutation, and aberration themselves,
show conclusively that Method 2 leads to smaller residual errors (Porter and Sadler,
1953; see also Section 3.344; also, it is more logical to apply aberration with respect
to a fixed frame of reference.)
Prom 1984 onward, the day numbers printed in The Astronomical Almanac are
calculated according to Method 2, and are designed for use with star places referred
to the mean epoch and equinox of the middle of the Julian year.
3.341 Methods Using Day Numbers The apparent right ascension and declination
(ai , 6 j ) , of a star at an epoch t+T, which includes the effects of precession, nutation,
annual aberration, proper motion, and annual parallax, and includes the second-
order day numbers (J and J' see Section 3.344), may be calculated either using the
3 / CELESTIAL REFERENCE SYSTEMS 1 57
Besselian day numbers A, B, C, D, E, and star constants (Section 3.342), from
+ Aa + Bb+Cc + Dd + E + J tan2 6 + r/x„ + w(dX - cY),
5]=6+Aa' + Bb' + Cc' + Dd' + f tan 6 + Tfis + ir(d'X - c'Y), (3.341-1)
or using the independent day numbers /, g, G, h, H (Section 3.343), from
qi = a + f + g sin (G + a) tan 6 + h sin (H + a) sec 6+ J tan2 <5
+ T/za + w(dX - cY),
S\ = 8 + g cos (G + a) + h cos (H + a) sin <5 + i cos 6 + f tan S
+ Tvs + K(d'X-c'Y), (3.341-2)
where a, 6 are the right ascension and declination of the star for the mean equinox
and epoch of f, and fia and fig are the annual proper motions in right ascension
and declination, respectively, and tt is the parallax. X and Y are the barycentric
coordinates of the Earth. The day numbers are usually tabulated in seconds of arc;
when used for calculating the star's right ascension, measured in time, either they
or the star constants by which they are multiplied should be divided by 15.
For stars, the corrections for the deflection of light, which has not been included,
are given in Section 3.261.
3.342 Besselian Day Numbers and the Star Constants To the first order, the Besselian
day numbers A, B, and E correct for precession and nutation; C and D correct for
annual aberration, and are given by
Atb
A =nr + n—— = tit + sin £q At/>,
ib
B - -Ae,
E = A'— f., (3.342-1)
where A' = xa(T = Q, t + r), tb' = ipA(T = 0S t + T% ftj and m (for the independent day
numbers) are the processional parameters (Section 3.211), and the nutation angles
Atb, Ae, and e0 the mean obliquity of the ecliptic, are given in Section 3.222. They
are all evaluated for the epoch t + T.
Since 1960 (see Section 3.53 for previous methods) the aberrational day num-
bers are a function of the Earth's barycentric velocity r, and are given by
D=-kX C=+kY I = +kZ,
(3.342-2)
158
EXPLANATORY SUPPLEMENT
where r' = (X, Y,Z) are the velocity components, in AU per day, referred to the mean
equinox of t (since aberration is applied first), and k = (t^/86400) x R x 3600, where
ta = 499. 004782 is the light- time for unit distance in seconds, and R converts from
radians to degrees (i.e., k converts from All/per day to seconds of arc).
If, however, the barycentric velocity of the Earth (ro) is referred to the standard
equinox of J2000.0, (e.g., as in DE200), then use the precession matrix P (see
Section 3.21) to precess these coordinates from the fixed equinox of J2000.0 to the
equinox of date t, i.e.,
r = P[J2000. 0, f] r0 (3.342-3)
The quantities Aip, Ae, A, B, C, D, E, in seconds of arc, and r are tabulated
daily in The Astronomical Almanac.
The star constants, which are constant only for the mean equinox of the fixed
epoch t, are defined by
a = m I n + sin a tan S, a' = cos a,
b = cos a tan 6, b' = - sin a,
(3.342-4)
c = cos a sec 8, c' - tan e0 cos 8 - sin a sin 8,
d — sin o. sec 8, d' = cos a sin 6,
where a and 8 are the mean place of the star for epoch f. The precession rates
(Section 3.211) m and n, and 6q, also should be evaluated for epoch f.
3.343 Independent Day Numbers The independent day numbers /, g, and G cor-
rect for precession and nutation, and h, H, and i correct for annual aberration. They
are given by
/ = (m/n)A +E = mr + cose At}>, hsinH = C,
g sin G^B, hcosH = D, (3.343-1)
gcosG = /l, <' = Ctane.
These day numbers are defined in terms of the Besselian day numbers, which are
given in Section 3.342.
;
3.344 Second-Order Day Numbers In the equations of Section 3.341 the terms
/ and J' are called the second-order day numbers, and approximate the second-
order terms that have been ignored. Table 3.344.1 lists all the second-order terms
neglected from approximations made at the various stages.
3 / CELESTIAL REFERENCE SYSTEMS
159
Table 3.344.1
Second-Order Terms
Term
Act cos S
AS
1
+/gcos (G + a) sin 6
-fg$m(G + a)
2
— ifocos Gcosasini5
+ ^/gcos Gsin a
3
+iy2sin2(G + (T)tan(5sin<5
-ig2 sin2 (G + a) tan 6
4
cos a sin (2G + a) cos 8
5
+ |h2sin2(H+ a) sec 6
-| ft2 sin2 (H + a)tan£
6
+^rr cos^ (H + a) sin 28
7
+h/cos(H + a) cos 28
8
sin 28
9
+gh sin (G+H+ 2a) tan S
-gh sin (G + a) sin (H + a) sec 8
10
+g/sin{G + a)
Terms 1 to 4 are neglected in the approximations for precession and nutation, 5
to 8 are neglected at the annual aberration stage, and 9 and 10 result from Method
2. Method 1, which is not considered here, gives rise to more terms.
Thus the second-order corrections / and /' are such that the corrections to right
ascension and declination are 7tan2<5 and J'tantf, respectively. They are derived
from the most significant terms, i.e., terms 3, 5, and 9, given in Table 3.344.1, and
replacing sec 8 by ± tan 8, with an error that vanishes at the poles. This gives
J = +[g sin (G+a)±h sin (H + a)] [g cos (G + a)±h cos (H + a)]
■ +[(A + D) sin a + (B±C) cos a] [(A ± D) cos a — (B±C) sin a],
J' = --|[gsin{G + a)±h sin (H + a)f
= -l-[(A±D) sin a + (B±Q cos a]2, (3.344-1)
the upper sign being taken for positive declinations, and the lower sign for negative
declinations. The expressions for second-order terms and J and / assume that /, g,
h are measured in radians, and J must be divided by 15 to be converted to seconds
of time.
These second-order day numbers J and f are tabulated in The Astronomical
Almanac as simple functions of north or south declinations, right ascension, and
date, so that the complete reduction may be made in one operation.
160
EXPLANATORY SUPPLEMENT
Table 3.344.2
Errors Due to Neglecting Second-Order Terms
0:007
0"008
0T009
0XJ10
0:012
0:014
o:oie
o:ois
0:020
0 < T < +1
5°
25°
35°
43°
49°
54°
57°
60°
63°
57°
62°
67°
70°
72°
74°
76°
78°
81°
The method is of advantage in the routine calculation of a number of star places.
In particular, Porter and Sadler (1953) clearly show the advantages of restricting
the range of r to ± 0. 5; this has the effect of reducing all the second-order terms
in / and g, which are functions of time. A more detailed analysis of the magnitude
of the neglected terms in different methods has confirmed the conclusion that if
t is allowed to reach +1, and no second-order corrections are applied, there are
unavoidable errors of O'.'OIO, even at declinations of 45°. If t is restricted to ± 0.5,
and the J and J' terms are included, then the maximum error reduces to 0"003, and
the range of declinations over which second-order corrections may be neglected is
correspondingly increased.
Table 3.344.2 gives the upper limit of declination for a given error when no
second-order terms are applied.
3.35 Topocentric-Place Algorithm
The topocentric place of a star or planet refers to its apparent direction as it would
actually be observed from some place on Earth, neglecting atmospheric refraction.
The apparent place, developed in Sections 3.31 and 3.32, can be thought of as the
apparent place of an object for a fictitious observer located at the center of a trans-
parent nonrefracting Earth. The difference between the apparent and topocentric
place is due to the slightly different position and velocity of an observer on the
Earth's surface compared with those of the fictitious observer at the Earth's center.
The change in direction of the observed body due to the difference of position is
referred to as geocentric parallax, and is significant only for objects in the solar sys-
tem. It is typically a few arcseconds for most solar-system bodies, but reaches about
1° for the Moon. The change in direction due to the difference in velocity (due to
the rotation of the Earth) is referred to as diurnal aberration, and is independent
of the distance of the observed body, and is always less than 0'.'32.
Atmospheric refraction also affects the apparent direction of celestial objects.
In fact, refraction at all wavelengths is orders of magnitude larger than either geo-
centric parallax or diurnal aberration. Refraction is discussed in Section 3.28, and
3 / CELESTIAL REFERENCE SYSTEMS 161
as it is usually considered a correction to observations rather than an effect to be
taken into account when computing a topocentric place, it is not considered here.
The simplest way of computing a topocentric place is to compute an apparent
place using the position and velocity vectors of the observer rather than the center
of mass of the Earth. That is, modify the barycentric vectors EB(f) and Eb(0 by
adding the position and velocity of the observer relative to the center of the Earth,
see Step h. This procedure is equally applicable to stars and solar-system objects.
The development that follows requires quantities related to precession and nutation
that, in the computation of geocentric apparent places, are not needed before Step
q. These quantities should be computed, used here, and saved for later use; they
are specifically noted as they arise.
3.351 Location and Universal Time of the Observation
Step a: Determine the universal time of observation, specifically, the epoch of
observation in the UT1 timescale. UT1 is affected by unpredictable irregularities
in the Earth's rotation, but is always within 0?9 of UTC, the latter defining civil
time and broadcast worldwide according to international convention. The difference
AUT = UT1 — UTC is determined and distributed by the International Earth
Rotation Service. The predicted value of AUT to within 011 (denoted DUT) is also
coded into UTC broadcasts. (See Section 2.24.)
Step b: Obtain r (in meters), the position vector of the observer in an Earth-
fixed, geocentric, right-handed coordinate system, with the jcy-plane the Earth's
equator, the jcz-plane the Greenwich meridian, and the z-axis pointed toward the
north terrestrial pole. In terms of the observer's geodetic latitude <f>, longitude A
(east longitudes positive), and height h above the Earth's reference ellipsoid [for
most purposes the height above mean sea level (the regional geoid) can be used], r
is given by
"(flC + A)cos<pcos A"
r= {aC + h) sin 4> sin \ , (3.351-1)
(aS + h) sin (p
where a is the equatorial radius of the Earth,
C = (cos <f>2 + ( 1 - f)2 sin2 <t>T "2, S = ( 1 - f)2C, (3.35 1-2)
and / is the adopted flattening of the Earth's reference ellipsoid. The IAU (1976)
constants are a = 6378 140 m and / = 1 / 298. 257.
162 EXPLANATORY SUPPLEMENT
3.352 Apparent Sidereal Time at the Epoch of Observation
Step c: Use Step s of Section 3.319 to obtain the two fundamental nutation angles,
Aip and Ae, and the mean (e0) and true (e) obliquity of the ecliptic. Save all these
quantities for later use.
Step d: Using the UT1 epoch of observation as the argument, compute the Green-
wich mean sidereal time 9m
8m = 67310!54841 + (876600h + 8640184?812866)7;
+ 0^093104 Tl - 6*2 x \0~6Tf,
(3.352-1)
where Tu is the number of centuries of 36525 days of universal time from 2000
January 1, 12h UT1 ( JD 2451545.0 UT1). The Greenwich apparent sidereal time is
then
9 = 8m + Atp cos e, (3.352-2)
where 9m and must be in the same units. Greenwich mean and apparent
sidereal time are the angles between the Greenwich meridian and the mean and
true equinoxes of date, respectively, and with the equation of the equinoxes (Sec-
tion 3.223} are tabulated daily at 0h UT1 in the The Astronomical Almanac.
3.353 Geocentric Position and Velocity Vectors of the Observer
Step e: Compute the geocentric position and velocity vectors of the observer, with
respect to the true equator and equinox of date, in meters, and meters/second,
respectively, from
Btt = R3(-0)Ri(y,)R2(*P)r,
g(f) = wk a g(f)
— sin 8 - cos 8 0
cos 6 — sin 9 0
0 0 0
Ri(y/,)R2(V,
(3.353-1)
(3.353-2)
where w = 7. 2921 151467 x 10~5 radians/second (Aoki et al, 1982) is the standard
value of the rotational angular velocity of the Earth, 6 is the Greenwich apparent
sidereal time at the time of observation, calculated in Step d, and k is a unit
vector pointing toward the north Celestial Ephemeris Pole of date. The angles xp,
yp correspond to the coordinates of the Celestial Ephemeris Pole with respect to
the terrestrial pole measured along the meridians at longitude 0" and 270° (90°
west) (see Section 3.27).
3 / CELESTIAL REFERENCE SYSTEMS 1 63
The preceding expressions take into account polar motion, which affects the
components of the observer's geocentric position vector at the 10-meter (0'.'3) level,
and may be neglected if desired by replacing TH\(yp) and RiCfp) by unit matrices. Ne-
glecting polar motion affects the computed topocentric place of the Moon by several
rnilliarcseconds, with a much smaller effect, inversely proportional to distance, for
other bodies. If effects at this level are important, corrections are also required to
refer the regional geoid (the coordinate system for the observer's geodetic latitude,
longitude, and height) to the Earth's reference ellipsoid. In forming the velocity of
the observer a standard value for the rotation rate of the Earth has been used, and
the small effects due to the variation in rate from this value and due to the change
of the polar motion components have been ignored.
More information on the computation of an observer's geocentric coordinates
are given in Section K of The Astronomical Almanac, Mueller (1969), Taff (1981),
as well as Chapter 4.
Step f: Convert the geocentric position and velocity vectors to units of AU and
Au/day, respectively, by multiplying g(/) by 1 1 A and g(t) by 86400 / A, where A =
1.49597870 x 10n is the number of meters in 1 au (IAU (1976)).
3.354 Position and Velocity of the Observer in the Space-Fixed Frame
Step g: Transform the vector g(/) and g(f) to the coordinate system defined by the
Earth's mean equator and equinox of the reference epoch to, which is the space-fixed
coordinate system in which the position and velocity of the Earth are expressed:
G(f) = P-'N"1 g(0 = PTNT g(f) = RT g(r),
G(f) = P-'N-1 g(f) = PTNTg(0 = RT g(r), (3.354-1)
where P and N are the precession and nutation matrices, developed in Steps q and
s in Sections 3.318 and 3.319. Here the inverse matrix is simply its transpose. The
most efficient procedure would be to evaluate the elements of these matrices at
this point and save them for later use in Steps r and t. Steps (q) and (s) could be
skipped.
Strictly, the precession and nutation matrices define a transformation between
a space-fixed system and a slowly rotating system. The slow rotation is the chang-
ing orientation of the Earth's axis due to external torques that the precession and
nutation theories describe. Therefore the conversion of the observer's velocity given
in Equation 3.354-1 is missing a Coriolis term. However, the equivalent linear ve-
locity of this rotation is of order 10-5 m/second for an observer on the surface of
164 EXPLANATORY SUPPLEMENT
the Earth, comparable to the tracking velocity of large telescopes and completely
negligible.
3.355 Barycentric Position and Velocity of the Observer
Step h: Calculate the barycentric position and velocity of the observer Ob(0 and
0B(f), by adding vectors G(f) and G(r), obtained above, representing the geocentric
position and velocity of the observer, to EB(f) and EB(r), obtained in Step f of
Section 3.31, which represents the barycentric position and velocity of the Earth,
thus
Ob = EBW + G(f),
OB = EB(0 + G(0. (3.355-1)
Then redefine the vectors EB(f) and EB(/) to be identical to OB(0 and 6B(/).
Step i: Continuing at Step g in Section 3.31, the other steps follow as before,
except that the elements of the precession and nutation matrices need not be re-
computed in Steps q and s.
The right ascension a and declination 6 obtained at Step u represent the
topocentric place of the object at the epoch of observation. The topocentric hour
angle of the object is given by LHA = 0 + X - a, where objects west of the meridian
(setting) have positive hour angles.
In many cases, the previous procedure may be simplified, but most care is
needed for objects in the inner solar system when the highest precision is required.
For objects beyond the inner solar system, to milliarcsecond precision, nutation
can be ignored throughout. That is, mean sidereal time can be used instead of
apparent sidereal time (the equation of the equinoxes can be considered zero) and
the nutation rotation matrix in Equations 3.354-1 can be neglected (i.e., N can be
considered the unit matrix). Additionally, the difference between UT1 and UTC
timescales can be ignored. However, these simplifications may not result, in any real
computational saving. The nutation parameters would have to be calculated anyway
in a later step. Furthermore, these simplifications affect the computed sidereal time
at the 1" level, and will therefore cause errors in the computed topocentric hour
angle of this magnitude. Therefore, in many cases, carrying out the full procedure,
and saving the values of the relevant nutation and time variables for later use, may
be the most prudent course.
3 / CELESTIAL REFERENCE SYSTEMS
3,4 DIFFERENTIAL ASTROMETRY
165
For differential astrometric measurement, the algorithms in Sections 3.31 and 3.32
can be simplified. In differential work it is necessary to consider only effects that
can alter the angles between the position vectors of the observed bodies, i.e., arc
lengths on the celestial sphere. The orientation of the coordinate system is not
considered of fundamental importance, since in most cases the celestial and instru-
mental coordinate systems are coupled. In any event, in differential observing, the
coordinate system is not established until after the fact, during the reduction of the
observations.
There are three types of differential positions: the virtual place, the local place,
and the astrometric place. For the reduction of high-precision differential observa-
tions, the local place should be used. With available computing power, it is straight-
forward to compute the local place of all objects within a field and use the ensemble
of local places as the starting point for the reduction procedure.
3.41 Virtual Place
The virtual place can be thought of as an apparent place expressed in the coordinate
system of the reference epoch f0- It represents the position of the star or planet as it
would be seen from the center of mass of the Earth at some date, in the coordinate
system defined by the Earth's mean equator and equinox of the reference epoch,
assuming that the Earth and its atmosphere were transparent and nonrefracting.
For the reduction of high-precision differential astrometric observations, there-
fore, the final precession and nutation rotations need not be performed, and Equa-
tions 3.31-1 and 3.32-1 reduce to, respectively,
u4(r') = /u?[uB(/ -t)- EB(/)]], (3.41-1)
and U4<'') = / lg[uB('o) + (' - 'o)UB(fo) - EB(0]L (3.41-2)
where all the symbols have been defined in Sections 3.31 and 3.32. The resulting
position is called the virtual place and is computed by following, in Section 3.31,
Steps a through p, then setting Oj = u2 and skipping to Step u, or equivalently for
stars, in Section 3.32, Steps 1 through 7, and Step 10 where 114 = u2- The omission of
the final precession and nutation rotations does not introduce any approximations
or distortions, since only orthogonal transformations are omitted.
3.42 Local Place
The local place is essentially the topocentric place expressed in the coordinate
system of the reference epoch f0. Local place is related to topocentric place in the
1 §6 EXPLANATORY SUPPLEMENT
same way that the virtual place is related to apparent place. Specifically, the local
place represents the position of a star or planet as it would be seen from a specific
location on Earth at some date and time, in the coordinate system defined by the
Earth's mean equator and equinox of the reference epoch, assuming the atmosphere
were nonrefracting. To compute it, simply add the procedure for calculating the
topocentric place (Section 3.35) to the procedure for computing the virtual place
(Section 3.41). (Note: The precession and nutation rotations in Section 3.354 should
not be omitted.) The local place has utility beyond its use in relative astrometry.
3.43 Astrometric Place
In differential work it has also been customary, if not strictly correct, to neglect
both the gravitational deflection of light (function g[...]) and the aberration of light
(function /[...]). The assumption is that for sufficiently small fields these differential
effects are so small that the relative positions of objects remain unaffected. Similarly
differential refraction is ignored, although its effect could be much larger because
of the variation in color over the same fields. Any residual distortion of the field
resulting from the neglect of these effects is assumed to be absorbed into plate
constants or similar parameters solved for in the data-reduction process.
The astrometric place of the planet or star is obtained from Equations 3.31-1
and 3.32-1, which reduce to
u4(/') = unit - <rj— EB(f) (3.43-1 )
and u4(f') = uB(f0) +(/ - fo)uB(fo) - EB(r), (3.43-2)
and is obtained by following, in Section 3.31, Steps a through 1, or equivalently for
stars, Steps 1 through 5, then setting u4 = U and skipping to the last step.
However, it should be recognized that the gravitational deflection of light should
not really be ignored in this way, since it cannot in principle be absorbed into plate
constants or similar reduction parameters: the deflection is a function of position
and distance. Although in any part of the sky the direction of the deflection is the
same for all bodies, its magnitude is less for solar-system bodies than for stars (see
Figure 3.26.2). Generally, this detail is of little practical importance, since only in
a few special cases can it cause errors exceeding 0"01.
Astrometric places are simple to compute, and therefore have been widely used.
Another attractive feature is that they can be directly plotted on an ordinary star
map with negligible error in the resulting field configuration. Astrometric places
are therefore used for the ephemerides of faint or fast-moving solar-system bodies,
such as minor planets and Pluto, in The Astronomical Almanac.
3 / CELESTIAL REFERENCE SYSTEMS 1 6/
For observations of solar-system bodies, it is frequently useful to compute a
topocentric astrometric place. Only the correction for geocentric parallax is ap-
plied. To compute a topocentric astrometric place, simply add the procedure given
in Section 3.35 to the procedure given above for computing an astrometric place.
(Again, the precession and nutation rotations in Section 3.354 should not be omit-
ted). In this case, however, the observer's velocity vector need not be computed,
since aberration is ignored. Topocentric astrometric places of stars are never re-
quired, since the topocentric correction is vanishingly small.
Before 1984, the effects of the E-terms of aberration were omitted; a detailed
discussion on them is given in Section 3.53. Also, in the past the term "astrographic"
place was used instead of astrometric place.
3.5 TRANSFORMATION TO FK5 SYSTEM AND EPOCH J2000.0
Expressions are given for converting both observational and compiled catalogs from
the old FK4 reference system of B1950.0 to the new FK5 reference system of
J2000.0. A detailed discussion of the formal background is also given.
3.51 FK4 Zero-Point Correction in Right Ascension
An early discussion of the problem of the position and motion of Newcomb's equinox
and the FK4 equinox can be found in Blackwell (1977). Later, it was determined by
Fricke (1982) that the FK4 right-ascension system requires a correction of +0?035
at B1950.0, This has been referred to as an equinox correction. It implies that the
right ascension of every star in any catalog whose right ascensions are referred to
the system of FK4 at epoch and equinox B1950.0 must be increased by 4O!035. The
motion of the equinox of +0?085 / cy. Fricke (1982) and the correction to lunisolar
precession of +l"10/cy Fricke (1977) implies that the FK4 system has a rotation
in right ascension relative to our present best realization of an inertial system of
(Of 085 — 1 '.' 1 0) per century. This has been corrected in the FK5 by altering the
proper motions by O!085/cy and precession by l'.'10/cy.
The equinox correction at any epoch, T, given as a function of its value at
B1950.0 and the difference in epoch (T - 7b), has been given by Fricke (1982) as
follows:
E(T) = 0!035 + 0!085(F - T0) 1 36524. 2198782, (3.51-1 )
where T is the Julian date at any epoch and To is the Julian date corresponding
to the beginning of the Besselian year B1950.0, i.e., 7*0 = 2433282.42345905. We
suppose that the divisor of the epoch difference should be the number of Julian
1 68 EXPLANATORY SUPPLEMENT
days per tropical century, but no significant error is made if the number of Julian
days per Julian century (36525) is adopted instead.
Aoki et al. (1983) have made a convincing argument that, for the sake of consis-
tency with the new definition of UT introduced at 0h UT on 1984 January 1 (Aoki
et al, 1982), it is important that the FK4 equinox correction should be applied to
right ascensions of the FK4 catalog referred to the same equinox and epoch (1984
January 1) as the date on which the new definition went into effect. The equinox
correction corresponding to that epoch is accordingly,
£(2445700. 5) = +0^06390. (3.5 1-2)
The reduction to the IAU 1976 system of constants at J2000.0 of the right-ascension
system of an observational catalog (as distinct from a compiled catalog) referred
to the FK4 system at an arbitrary equinox and at the mean epoch of observation
should be done in a manner consistent with the way in which the catalog was
referred to the system of FK4 at the time of observation. On the other hand, if an
observational catalog has been constructed from a fundamental treatment of the
observations of solar-system objects and thus referred to the observed dynamical
equinox, the reduction to the IAU 1976 system of constants at J2000.0 would ignore
the zero-point correction normally required by the FK4 system in right ascension.
3.52 The Correction to the FK4 Proper Motion System in Right Ascension
The existence of a secular term of the form 0?085(r - T0) I 36524.2198782 in the
equinox correction has generally been interpreted as a constant, nonprecession-
dependent correction applicable to the FK4 right-ascension proper-motion system
(Fricke 1982). This implies that at any epoch, and regardless of the equinox, all
right-ascension proper motions of the FK4 catalog (or any catalog with the proper
motions referred to the FK4 system) must be increased by +0!085 per tropical
century.
The nonlinear character of precession viewed as a mathematical operation sug-
gests that subtle differences can be introduced into a catalog transformation from
one epoch and equinox to another, depending on the epoch at which corrections
are introduced. A number of possibilities come to mind. One could, for example,
apply the corrections to positions and proper motions referred to:
(1) the epoch and equinox of the beginning of 1984; or
(2) the epoch and equinox of B1950.0; or
(3) the epoch and equinox of the mean epoch of observation.
Among these three options, the third one most closely approximates the op-
eration that would have been carried out if the improved IAU 1976 system had
3 / CELESTIAL REFERENCE SYSTEMS 1 69
been available at the time of observation. In the case of individual observational
catalogs, such as the series of fundamentally observed Washington Six-inch Transit
Circle catalogs, the reduction to the FK5 equinox is best done by directly calculat-
ing the difference FK5-Catalog from the stars in common to both catalogs.
3.53 Elliptic Terms in Aberration
3,531 Effect on Position Recommendation 4 of IAU 1976 Resolution No. 1 indi-
cates that from 1984 onward, stellar aberration is to be computed from the total
velocity of the Earth referred to the barycenter of the solar system, and mean
catalog places are not to contain elliptic terms of aberration.
The problem of reducing observational catalogs to a uniform system is compli-
cated by the change in the conventional value of the constant of aberration, which
since 1911 had been taken as 20'.'47 (Paris conference, 1911), and from 1968 has been
taken as 20'.'496 (IAU 1967). Before 1911, the value of Struve and Peters, 20"4551,
was in common use, and from 1984 the value of 20'.'49552 has been introduced as
the conventional value.
Reduction to a uniform system is further complicated by changes in the method
of computing the aberrational day numbers, C and D, given in the national ephem-
erides and almanacs. Before 1960, the aberrational day numbers C and D were
calculated from a circular approximation to the motion of the Earth in which small
periodic terms due to the action of the Moon and planets and the elliptic terms in
aberration were neglected (see Explanatory Supplement, p. 48ff.).
Beginning with 1960, the aberrational day numbers C and D were derived
from the true velocity of the Earth referred to the center of mass of the solar
system and to a dynamically determined frame of reference. The elliptic terms in
aberration due to the eccentricity of the Earth's orbit were then removed from an
otherwise completely correct expression of the annual aberration. Furthermore, as of
1960, ephemeris values of the aberrational day numbers properly reflect the slowly
changing eccentricity and longitude of perihelion of the Earth's orbit (Explanatory
Supplement, p. 158ff.).
When one is working with individual observational catalogs, reductions to a
uniform system of annual aberration should take into account the changes in the
value of the constant of aberration and the method of calculation of the published
values of the aberrational day numbers, C and D. This will adjust the published
values of the observed positions in a manner consistent with conventions in use at
the time the observations were made and reduced from apparent to mean place.
If this procedure is not followed, then corrupted data will be produced, and the
potential for systematic improvement will be diminished.
The FK4 catalog, and other fundamental catalogs such as the N30 (Morgan,
1952) and the General Catalog (Boss 1937), were compiled from observations made
prior to 1960, when a circular approximation was made to the motion of the Earth
1 70 EXPLANATORY SUPPLEMENT
in its orbit, and the elliptic terms in aberration were ignored, except for stars within
10° of the poles in the case of the FK4 catalog.
Due to the rapid, daily change in the value of the circular component of an-
nual aberration, it is not possible to correct mean positions for changes that have
occurred in the adopted value of the constant of aberration either in the compiled
fundamental catalogs or in observational catalogs of mean positions. Observations
of solar-system objects are normally given on a daily basis and may be corrected
with a high degree of rigor. However, because the elliptic terms in aberration change
so very slowly, their influence can be removed from the FK4 catalog positions by
subtracting the elliptic terms in aberration from the catalog right ascension and
declination (aeat, 6cat) to obtain the corrected right ascension and declination (a, 6)
as follows:
a = acat - (ACcos acM + AD sin mm) I (15 cos 4at),
<5 = 6cat - ( AD cos acat - AC sin acat) sin <5cat - ACtan ecos 6cat, (3.531-1)
where AC = -0"065838, AD = +0:'335299, and ACtane = -01'028553 at epoch
B1950.0, using the J2000.0 value of the constant of aberration.
These equations represent the classical elliptic aberration with first-order ac-
curacy in the ratio of the Earth's velocity to the velocity of light (Woolard and
Clemence 1966, p. 113ff). More than the necessary number of significant figures is
carried in order to ensure correct rounding.
For applications requiring accuracies better than 1 milliarcsecond (mas), the
discussions of Stumpff (1979, 1980a) regarding the second-order relativistic terms
in elliptic aberration should be consulted.
The numerical values for AC, AD, and ACtane given here are not,, generally
applicable to every case in which positions have been referred to the equinox of
B1950.0. For an observational catalog referred to an arbitrary equinox, each position
(where right ascension and decimation must be treated separately if their mean
epochs of observation are not the same) may be precessed to the equinox of the mean
epoch of observation and then corrected for the elliptic terms of aberration using
values referred to the moving equinox (see, e.g., Scott 1964). Many observational
catalogs will require an epoch-dependent correction of this type. Compiled catalogs
may or may not require a correction, depending on how the compilation was done
(Lederle and Schwan, 1984).
The time- dependent expressions for the coefficients of Equation 3.531-1 may
be written as follows (Woolard and Clemence, 1966, p. 114ff):
AC = — ke cos F cos e,
AD = — kesinT,
(3.531-2)
3 / CELESTIAL REFERENCE SYSTEMS 1 71
wnere k = 20'.'49552 is the constant of aberration at J2000.0,
e - Eccentricity of the Earth's orbit
a 0. 0167301 1 - 0. 00004193(7 - 7b) - 0. 000000126(7 - 7b)2,
T = Mean longitude of perigee of the solar orbit
= 282°04'49,.,951 + 619Q':67(7 - 7b) + 1"65(7 - T0f + 0'.'012(r - 70)3,
e = Obliquity of the ecliptic
= 23°26'44:'836 - 46'.'8495(r - T0) - 0'.'00319(7 - T0f + 0'.'00181(7 - 7b)3,
where (T - To) is in units of Julian centuries of 36525 days and 7b is the epoch
B1950.0 (2433282.42345905). Equation 3.531-2 has been developed from ES, p. 98,
and may be used to refer Equation 3.531-1 to any epoch of observation relative to
the moving (i.e., of date) equinox.
For recently observed Washington catalogs, such as the W5(50) Six-inch .Tran-
sit Circle Catalog (1963-1971) (Hughes and Scott 1982) and the WL(50) Seven-inch
Transit Circle Catalog (1967-1973) (Hughes, Smith, and Branham, in press) ob-
served from El Leoncito in San Juan, Argentina, the assumed constant of aberration
used in the apparent place computations was 20"496. The W6(50) Six-inch Transit
Circle Catalog (1973-1982) (currently under discussion) will be re-reduced in strict
accordance with the IAU 1976 resolution, and will be renamed to reflect J2000.0 as
the equinox to which it will be referred.
3.532 Effect on Proper Motion Aoki et al. (1983), introduce a correction to the
proper motions to compensate for a secular change of the elliptic aberration asso-
ciated with the variation of the eccentricity of the Earth's orbit, of the longitude of
solar perigee, and of the obliquity of the ecliptic with time. Lederle and Schwan re-
mark that this is unnecessary because of the practice followed in the compilation of
FK4 proper motions by which differences in right ascension and declination between
observed positions and positions computed on the basis of the FK3 were used. In
fact, the problem is complicated by the practice of having eliminated the influence
of elliptic aberration for circumpolar stars before precessing to a new equinox and
epoch, and then re-introducing the effect of elliptic aberration appropriate for the
new equinox and epoch. This was not done for the stars between +80° and -80°
declination (FK3, 1937; FK4, 1963), which means that no single practice will work
equally well or consistently for all stars of the FK4, nor for any catalogs referred to
the FK4 system. Corrections for the secular change of elliptic terms in aberration
are not necessary for proper motions of circumpolar FK4 stars, but a correction is
required for FK4 stars between +80° and —80°.
A very good case can be made that the highest-accuracy transformation of
the FK4 catalog to the equinox of J2000.0 would result if the procedure outlined
1 ' ' EXPLANATORY SUPPLEMENT
above were followed paying special attention to the treatment of the stars as regards
the elliptic terms in aberration. Stars within 10° of the poles need have only the
elliptic terms in aberration at B1950.0 removed from their positions. Their proper
motions need no modification. For stars between +80° and -80° in declination, the
positions and proper motions should be precessed to the equinox of the mean epoch
of observation and the elliptic terms in aberration (referred to the moving equinox
of the mean epoch of observation given by Equation 3.531-2) should be removed
from the positions. Equation 3.531-1 may be differentiated with respect to time to
give the equations for the corrected proper motions (p,pJ) in right ascension and
declination in terms of the catalog proper motions in right ascension and declination
(/^cacMcat)' and tne catalog positions (aca[,8cat),
/' = Urn - (ACcos c*ca, + AD sin o^,) / ( 15 cos <5cat) (3.532-1 )
- ^caI sin 1 "(-AC sin acaI + AD cos acal) / cos Sm
-/4t sin '"(ACcos Qcal + AD sin Qcal) tan 5ca[/(15 cos<5cat),
= /4at - (ADcosocat - ACsin acal) sin 6^ (3.532-2)
- 15/ica,sin l"(-ADsinocaI - ACcos acal) sin Scat
- Mcai sin l"(ADcosQcat - ACsin acal) cos <5ca[
- (ACtan e + ACef sec2 e) cos 6cal
+ 1 5 ficat sin 1 "ACtan e sin 6cst,
where the factor 15 converts arc to time or vice versa, and
AC = +ket r cos e sin T - k(e cos e - eer sin e) cos f, (3.532-3)
AD s —ke sin T — ketr cos T,
AC tan e = +ketr sin e sin T - k(e cos e - eir sin e) cos T tan e,
where fr and lr are the rates of change of T and e in radians per tropical century.
The quantities AC, AD, ACtan e have to be evaluated in a fixed frame at
B1950.0. To convert from the rotating frame of date to this fixed frame, the in-
stantaneous rates of change mo, no, and p0 at B1950.0 have to be removed from the
displacements in right ascension (f cose), declination (f sine), and longitude (f) due
to precession of the mean longitude of perigee.
3 / CELESTIAL REFERENCE SYSTEMS 1 73
Hence with
= -0.00004193 per tropical cy
mo = 4609"90 per tropical cy
f = — r|j-=o = +6190"54 per tropical cy «o = 2004"26 per tropical cy
dT
j po = 5026'.'75 per tropical cy
i = — f |r=o = -46'.'8485 per tropical cv
dT
(3.532^)
the numerical values for AC, AD, and ACtan e in the fixed frame at B1950.0 are
AC = -0"001580 per tropical century,
AD = -0'.'001245 per tropical century,
ACtan e = -0'.'000677 per tropical century. (3.532-5)
The equations for mo and «o are given in Section 3.214, and po = mo cos e + no sin e.
3.54 Precession
A detailed discussion on precession is given in Section 3.21. The new IAU 1976
precession angles are defined in Section 3.211, and the old Newcomb angles are
discussed in Section 3.214.
3.541 Newcomb's Precession The precessional motion of the reference frame from
the equinox and equator of B1950.0 to 1984 Jan l"(f is calculated using Andoyer's
Equation 3.214-1. The adopted values are tabulated in the last column of the Ta-
ble 3.214.1. These values were obtained with the following time arguments; t\ - 0. 1,
| = 0. 133999566814, and t = 0.033999566814. The resulting precession matrix is
" +0.999965667560
+0.007599409535
+0.003303433846
-0.007599409538
+0.999971123992
-0.000012551554
-0.003303433841"
-0.000012553023
+0. 999994543569
3.542 The IAU 1976 Precession The precessional motion of the reference frame
from the equinox and equator of 1984 Jan ld0h to J2000.0 is calculated using the IAU
(1976) precession angles defined by Lieske (1979) and and given in Section 3.211.
The values obtained by evaluating these expressions with T = -0. 1 600 1 36893 and
1 74 EXPLANATORY SUPPLEMENT
i s -T, and the accumulated angles from 1984 January l^O11 to J2000.U are QA =
368'.'9985, zA = 369"0188, and 9A = 320"7279, and the resulting precession matrix is
+0.999992390029
+0. 003577999042
+0. 001554929624
-0.003577999042
+0. 999993598937
-0. 000002781702
-0. 001554929623"
-0.000002781855
+0.999998791092
3.55 The Proper-Motion System
The discontinuity in the right-ascension proper-motion system of the FK4 catalog
is to be imposed at 1984 January ld0h. Proper motions referred to that equinox and
epoch should:
(1) be corrected for the zero point of the right- ascension proper-motion system
as discussed in Section 3.52,
(2) be changed from the Newcomb to the IAU 1976 (Lieske et al, 1977) pre-
cession basis, and
(3) be expressed in units of Julian rather than tropical centuries.
Only points 2 and 3 apply to the proper motions in declination.
The fundamental condition to be satisfied at 1984 January ld0h is that the
centennial variation on the new basis shall equal the centennial variation on the old
basis plus a constant correction in the case of the right-ascension proper motions:
^new — ^Vold +
finew = field- (3.55-1)
Let quantities without subscripts be taken as the new values referred to the IAU
1976 basis, and those with subscripts 'o' be taken as referring to the old (Newcomb)
basis, then in right ascension
fj, + m + « sin a tan 6 =
and in declination
ft' + n cos a -
(ft0 + m0 + no sin a0 tan 6q + E) F,
(fjt'Q + n0 cos oq) F,
(3.55-2)
(3.55-3)
3 / CELESTIAL REFERENCE SYSTEMS
where
175
are the proper motions in right ascension and declination, respec-
tively, referred to the IAU 1976 basis and precessed to 1984 January
ld0h;
are the centennial general precession in right ascension and decli-
nation, respectively, referred to the IAU 1976 basis at 1984 January
ld0h;
are the right ascension and declination for the FK4 catalog pre-
cessed to 1984 Jan 1;
= a0 + 0!06390;
= fio;
_ 0!085 per tropical century;
are the proper motions in right ascension and declination, respec-
tively, from the FK4 catalog, for example, and precessed to 1984
January l"0h;
are the centennial general precession in right ascension and decli-
nation, respectively, referred to Newcomb's precession at 1984 Jan-
uary ld0h;
is the factor for converting from tropical centuries to Julian
centuries.
The quantities m and n are calculated using the IAU 1976 precession (Sec-
tion 3.211) and at 1984 January ld0h with T = -0. 1600136893
m = 46H'.'98926 / Julian century " 307M65950 / Julian century, and
n = 2004"44743 / Julian century = 133!629829 / Julian century. (3.55^1)
Similarly mi0 and n0 may be derived from the equations for Newcomb's preces-
sion in Section 3.214, and at 1984 January ld0h with t\ =0. 13399956681,
m0 = 4610'.'95218 / Julian century = 307^396812 / Julian century, and
n0 = 2004'.'01 126 / Julian century = 133*600750 / Julian century. (3.55-5)
The values of m, n, mo, and n0 will be used later in Section 3.566.
3.56 Equations for the Transformation of Catalogs from B1 950.0 to J2000.0
This transformation is conversion from a mean place to mean place. That is, it trans-
forms catalog mean places and proper motions from one reference epoch (B1950.0)
to another (J2000.0). The complexity of the transformation results from the changes
Mi M'
m, n
ao, 8q
a
6
E
M0> Mo
mo, no
F
1 76 EXPLANATORY SUPPLEMENT
in constants, timescales, and procedures mandated by the IAU for epoch J2000.0
catalog data.
3.561 Units and the System of Positions and Proper Motions It is assumed that
the positions and proper motions of the catalog to be transformed to J2000.0 are
referred to the epoch and equinox of B1950.0. The positions are assumed to be
in units of seconds of time in right ascension and seconds of arc in declination.
However, when these quantities are used as arguments of trigonometric functions,
it is left to the user to express them in degrees or radians as necessary. Similarly, the
inverse of a trigonometric function is assumed to be degrees. The proper motions
are assumed to be in units of seconds of time per tropical century in right ascension
and seconds of arc per tropical century in declination. Radial velocities are taken
to be in units of km/s and the parallaxes are in seconds of arc. It is also assumed
that the catalog in question is referred to the system of the FK4. If not, steps must
be taken either to refer the catalog to the FK4 system or to alter the appropriate
steps in the discussion which follows.
3.562 Elliptic Aberration Correct the catalog right ascension and declination (acat.
<5Cat) for the elliptic terms in aberration using Equations 3.531 1 and 3.531-2. If
necessary correct the catalog proper motions for the elliptic terms in
aberration using Equations 3.532-1 and 3.532-2.
3.563 Position and Velocity Vectors (B1950.0 to 1984 Jan ld0h) At the equinox and
epoch of B1950.0 form the position vector m, and the velocity vector u, from the
corrected positions a and 6, the proper motions fi and pi (corrected if necessary),
the parallax p, and the radial velocity r, using Equations 3.23 1 and 3.23-2 given in
Section 3.23. Note that to obtain the required units of au/ tropical century s is the
conversion factor from seconds of arc to radians and k = 86400 x 36524.2198782 x
1.49597870 x 10-8 is the conversion from km/s to AU per tropical century. Thus
the unit of u, is astronomical units (au), and Ui is in AU per tropical century.
The position vector at epoch /, 1984 Jan l"0h (2445700.5), but still referred to
the equinox of B1950.0, is
u2(fi) = ui(/ti) + fin(fi - *o), (3.563-1)
where (/, - f0) js the number of tropical centuries between epoch t0 (B1950.0) and
epoch f| .
3 / CELESTIAL REFERENCE SYSTEMS 1 77
3.564 Precession from B1950.0to 1984 Jan ld0h Apply the precession from B 1950.0
to 1984 Jan ld0h to the vectors u2(fi) and U|, using the precession matrix P, calcu-
lated from the angles z, and 6 given in Section 3.541. Then
u3(f,) = PiU2(fi) (3.564-1)
and
u2 = Piii]. (3.564-2)
From the components *3, y3, z3, and x2, y2, Zz, of the vectors u3 and u2, respectively,
calculate the right ascension and declination and their proper motions at 1984 Jan
ld0h:
q, =tan-'0'3/.r3), (3.564-3)
6t = tan 1 (z3!ij4 + ?3 ) , (3.564-4)
(*3 V2 - V3*2)
I5s(4+j|)'
rife - z3(x3x2 +ym + z3z2)
15s(jc| + yj)
= wj^^t^ (3.564-6)
where r\ = x\ +y\ +z\.
The equations for /x2 and fi'2 follow directly from the total time derivative of the
equations for Q| and 6\. The units of a\ and 6\ are degrees; n2 and fi'2 are in seconds
of time per tropical century, and seconds of arc per tropical century, respectively,
with s denned in Section 3.563.
The parallax, p3, at 1984 Jan ld0h, consistent with a model assuming linear
space motion, is given by
p3 =sin-'(l In) (3.564-7)
and the new radial velocity in AU per tropical century, r3, is
r3 = (X3X2 +y3y2 + Ziii) I r3. (3.564-8)
3.565 Right Ascension Zero-Point Correction Apply the zero-point correction to
the right ascension:
a2 = a 1 +0!06390,
(3.565-1)
1 78 EXPLANATORY SUPPLEMENT
where the right ascension and declination (au 6{) are expressed in seconds of time
and seconds of arc, respectively.
3.566 Transformation of the Proper Motion System Applying the corrections and
transformations discussed in section 3. 55 to the proper motions:
M = (M2 + 0- 085) F - (m - m0) - (n sin a2 - ra0 sin at ) tan 62
= (m + 0. 085) F - (307. 465950 - 307. 3968 12)
- (133. 629829 sin a2 - 133. 600750 sin a,) tan52,
fi'3 = /i2 F - (n cos a2 - no cos a, )
= n'2F- (2004. 44743 cos a2 - 2004. 01 126 cos aj )
where F = 1. 00002135903. (3.566-1)
From Equation 3.566-1 it becomes evident that a star precisely at or within a
few seconds of arc of a pole must be treated as a special case. Historically, proper
motions have been given in right ascension and declination without problems. If
a star is very close to the pole, the singularity is avoided by considering the real
displacement ^3 cos 62 .
3.567 Position and Velocity Vectors (1984 Jan ld0h to J2000.0) Form the vectors
U4 and u4 from oc2, 62, ^3, ^3, using Equations 3.23-1 and 3.23-3 ensuring that the
unit of U4 is astronomical units (au), and ii4 is in AU per Julian century.
Form the position vector u5 at epoch J2000.0:
uste) = u4(l 1 ) + ii4(f2 - h ), (3.567-1 )
where (t2 - tL) is the number of Julian centuries between epoch t2 (J2000.0) and
epoch fj, 1984 Jan W.
3.568 Precession from 1984 Jan ld0H to J2000.0 Apply the precession from 1984
Jan ld0h to J2000.0 to the vectors u5 and 114, using the precession matrix P2 calcu-
lated from the angles &, and dA given in Section 3.542
u6 = P2 u5(*2); (3.568-1)
«5 = P2 u4. (3.568-2)
From the components of the vectors u6 and u5, the right ascension and declination
and their proper motions at J2000.0 along with the modified parallax and radial
velocity may be computed from a reapplication of Equations 3.564-3 to 3.564-8.
3 / CELESTIAL REFERENCE SYSTEMS
179
The radial velocity will be in AU per Julian century, and the conversion factor to
km/s is 1.49597870 x 108 / (86400 x 36525).
If neither the radial velocity nor the parallax of a star were known at the
beginning of this procedure, then the fictitious values produced by the procedure
should be ignored. If only the parallax is known, then the original B1950.0 value
should be brought forward without modification.
3.57 Transformation of Observational Catalogs
If the transformation is applied to an observational catalog in which no proper
motions are given, and the mean positions are referred to the mean epoch of obser-
vation and the equinox of B1950.0, then the mean positions should be precessed to
the equinox of the mean epoch of observation, which, in general, will be different
for each star. The right ascension and declination should be corrected foT terms in
elliptic aberration corresponding to the mean epoch of observation, provided they
had not already been removed when deriving the catalog place. Special care must
be taken if the epochs of observation in right ascension and declination are different,
which happens frequently in modern observational catalogs.
The right ascensions, if they were referred to the FK4 system, should be cor-
rected for the zero-point error evaluated at the mean epoch of observation in right
ascension. The positions may be brought forward to J2000.0 using the equations for
the IAU 1976 precession angles Ct, zA, and 6A given in Table 3.211.1 evaluated for
the time interval from the mean equinox of observation to the equinox of J2000.0.
In this case, the proper motion part of the computation may be ignored.
The same is true of catalogs of radio-interferometrically determined positions
of quasars and other objects presumed to be located at extragalactic distances.
However, the transformation of Very Long Baseline Interferometry (VLBI) and
connected-element interferometry (CEI) catalog positions from the equinox of
B1950.0 to J2000.0 is slightly more complicated than the transformation of the
FK4 catalog because of the way the zero point in right ascension is established.
This usually involves a systematic adjustment to all right ascensions. Most adjust-
ments refer to the averaged optical and radio right ascension of 3C 273B given by
Hazard et al. (1971). To the extent that this right ascension was referred to the
FK4 system at its mean epoch of observation, it also requires a correction to right
ascension by an amount that the FK4 equinox required at the same epoch. ^
Another factor that should be taken into consideration when transforming in-
terferometrically observed positions from the B1950.0 basis to the J2000.0 is the
intrinsically high accuracy of such a catalog. In order to preserve the high accuracy
of such catalogs, the effects that have been discussed earlier— the elliptic terms in
aberration, the equinox correction, and the use of Newcomb's precession constant
to refer observations to a fixed equinox— should all be removed in a manner as
closely analogous to the operations which would have been employed in reducing
1 80 EXPLANATORY SUPPLEMENT
the original observations at the epoch of observation had the improved constants
been available at that time. In genera], this means that catalog positions must be
precessed back to the equinox of the mean epoch of observation using the same
precession constants as the catalog compiler used, and if necessary corrected for
the elliptic terms in aberration and equinox error of the FK4 system. The modified
position is then precessed to the equinox of J2000.0 using the Lieske et al. pre-
cession. This procedure does not recognize 1984 as an epoch at which any special
condition is to be imposed and best preserves without distortion the high precision
of interferometrically determined positions.
3.58 Numerical Examples
Numerical examples are given in Table 3.58.1 for selected FK4 stars. The selection
was made on the basis of proximity to the north and south poles, the magnitude
of the proper motion, parallax and radial velocity, and one star which crosses the
equator between B1950.0 and ,12000.0. For each star, the first line gives the B1950.0
position on the FK4 system and the second line gives the transformed data for
J2000.0. The transformed FK4 data given in Table 3.58.1 are not the same as the
FK5 data resulting from the incorporation of new observational results produced
since the publication of the FK4 catalog.
3.59 Matrix Method
There has been much controversy in the literature over the correct procedure for
transforming catalog positions and proper motions to the FK5 system discussed
in the preceding sections (Smith et al., 1989). The matrix method described here
(Yallop et al, 1989) is based on Standish (1982b) but modified to agree with the
procedure given in Section 3.56. The transformation is categorized into nine steps;
six of the steps of the transformation are represented by a six-by-six matrix de-
scribed below. Sections 3.591 and 3.592 give the complete algorithms for converting
star places from B1950.0 to J2000.0, and from J2000.0 to B1950.0 by reversing the
transformation. It should be noted that when transferring individual observations,
as opposed to a catalog mean place, the safest method is to transform the obser-
vation back to the epoch of the observation, on the FK4 system (or in the system
that was used to produce the observed mean place), convert to the FK5 system,
and transform to the epoch and equinox of J2000.0.
The transformation for a fundamental catalog position is as follows:
Step 1: Form the position and velocity vector from the FK4 catalog data; i.e.,
form the position and velocity vector from the right ascension, declination, proper
motions, parallax, and radial velocity.
3 / CELESTIAL REFERENCE SYSTEMS
181
Table 3.58.1
Selected Star Positions on FK4 and FK5 Systems
FK4
a
6
Px
RV
No.
h
m
S
0
t
tt
s/cy
"/cy
km/s
10
00
17
28.774
-65
10
06.70
+27.141
+116.74
0.134
+ 8.70
10
00
20
04.3100
-64
52
29.332
+26.8649
+116.285
0.1340
+ 8.74
1 1 3
03
17
55.847
—43
15
35.74
+27.827
+ 74.76
0.156
+Afi Rn
119
03
19
55.6785
-43
04
10.830
+27.7694
+ 73.050
0.1559
+86.87
06
1 1
43.975
-74
44
12.46
+3.105
- 21.12
0.1 1 5
+35.00
239
06
10
14.5196
-74
45
1 1 .036
+3.1310
- 21.304
0.1150
+35.00
1 4
36
1 1 .250
-60
37
48.85
—49.042
+ 71.20
0.751
—22.20
538
14
39
36.1869
-60
50
07.393
-49.5060
+ 69.934
0.7516
-22.18
21
04
\J £7. JOU
+38
29
59.10
+35.227
+318.47
0.292
— R4 on
793
21
06
54.5901
+38
44
44.969
+35.3528
+320.206
0.2923
-63.89
01
48
48.784
+89
01
43.74
+18.107
- 0.43
0.000
+ 0.00
907
02
31
49.8131
+89
15
50.661
+21.7272
- 1.571
0.0000
+ 0.00
923
20
15
03.004
-89
08
18.48
+1 1 .702
0.09
0.000
+ 0.00
923
21
08
46.0652
-88
57
23.667
+8.4469
+ 0.171
0.0000
+ 0.00
1307
11
50
06.172
+38
04
39.15
+33.873
-580.57
0.116
-98.30
1307
11
52
58.7461
+37
43
07.456
+33.7156
-581.216
0.1161
-97.81
1393
14
54
59.224
+00
01
58.08
+0.41 1
- 2.73
0.000
+ 0.00
1393
14
57
33.2650
-00
10
03.240
+0.4273
- 2.402
0.0000
+ 0.00
Step 2: Remove the E-terms of aberration from the B1950.0 catalog mean place.
There is also a question whether the E-terms should be removed from the proper
motions or not. The problem is discussed more fully in Section 3.53 and the corre-
sponding vector development is given in Yallop et al. (1989). If the FK4 catalog is
used, they certainly do not have to be removed from stars within 10° of the poles,
because they have not been included (Lederle, 1984).
Step 3: Apply space motion to the position vector to the epoch 1984 January 1.0,
which is the epoch at which the sidereal time expression in terms of UT is changed
(IAU, 1976).
Step 4: Precess, using the FK4 precession constants, the position, and velocity
from B1950.0 to 1984 January 1.0.
Step 5: Apply the equinox correction FK4 to FK5 at 1984 January 1.0 to the right
ascension and proper motion in right ascension.
Step 6: Convert the proper motions from seconds of arc per tropical century to
seconds of arc per Julian century.
182
EXPLANATORY SUPPLEMENT
Step 7: Precess, using the FK5 precession constants, the position and velocity
from 1984 January 1.0 to J2000.0.
Step 8: Apply space motion to the position vector from 1984 January 1.0 to
J2000.0.
Step 9: Convert the position and velocity vector back to right ascension and dec-
lination and proper motions in right ascension and declination, and extract the
parallax and radial velocity in the FK5 system.
The Steps 3-8 may be represented by successive multiplication by the matrices Mj ,
M2, M3, M4, M5, and Me on the position and velocity vector at B 1950.0 to produce
the position and velocity vector at J2000.0 where
Y
= M
[X]
r
= M6 M5 M4 M3 M2 M|
(3.59-1)
and the position and velocity vectors r and r are expressed in AU and seconds of
arc per century, tropical or Julian, as appropriate. The conversion of the units of
the velocity vector is done by matrix Mi. The following is a brief explanation of
the notation used:
CB = 36524.21987817305 days, the length of the tropical century;
Cj = 36525 days, the length of the Julian century;
£ = Epoch, e.g., 1984 January 1.0;
JD(date) = Julian date, JD(B1950.0) = 2433282. 42345905, JD(J2000.0) = 2451545. 0,
and JD(1984 January 1.0) = 2445700.5;
I and O - Unit and null 3x3 matrices.
The rotation matrices M2, M3, and M5 are formed from the Q„ t = 2,3 matrices,
which are given by
Q,(4>, 0) =
R,(0) O
a
(3.59-2)
L^R,-(0) R,-(0).
where R,'s are the standard 3x3 rotation matrices, and ^ is a function of time.
The derivatives of the required rotation matrices are
— sin <^ 0
0 0
cos0 0
-COS(j>
0
- sin <p
d<i>
R3 W =
— sin*
— cos (
0
cos<6 0"
-sin0 0
0 0
(3.59-3)
In 6-space the notation for the precession matrix (Section 3.21) has to be mod-
ified to include a further parameter s, where s = 0 when precessing from one inertial
frame to another inertial frame, and s = 1 when precessing from an inertial frame
3 / CELESTIAL REFERENCE SYSTEMS
183
to a noninertial frame (i.e., rotating frame of date). The expression for P, using the
accumulated precession angle notation, becomes
P[£F, eD, s] =
P O
sP P
= Q3(-^,-^a)Q2(+^,+40Q3(-C4,-<4). (3.59-4)
In order to calculate P„ and P„ the equatorial precession angles on the FK4
(Section 3.214) and FK5 (Section 3.211) systems are required, respectively, together
with their rates of change with respect to time. The precession rates on the FK4
and FK5 systems are given by
QA = jtUT,t) zA = jtZA(T,t) eA = jt6A(T,t).
(3.59-5)
(3.59-6)
The six matrices are defined as follows:
M| Adds space motion between the standard epoch B1950.0 and e to the
position vector at B1950.0:
M, =
I f0I
O I
(3.59-7)
where t0 = c(JD(e) - JD(B1950. 0)) / CB and c = tt / (180 x 3600) is the factor that
converts the proper motions to radians per century.
M2 Applies FK4 precession from B1950.0 to g, to the position and velocity
in 6-space:
M2 = Po[B1950. 0, e, 1] = Q3(-z, -fl Q2(+#, +0) Q3(-Co, -Co)- (3.59-8)
M3 Adds the equinox correction to the right ascension and proper motion in
right ascension at epoch e:
M3 =Q3(-£e,-£),
(3.59-9)
where Ee = £50 + E(JD(e) - JD(B1950. 0)) / CB, £50 = 0'.'525 and E = 1'.'275.
M4 Converts the proper motions from tropical centuries to Julian centuries:
M4 =
I O
O (Q/CB)I
(3.59-10)
M5 Applies FK5 precession from e to J2000.0, to the position and velocity
in 6-space:
184
M5 =P-1[J2000.0,£,1]
EXPLANATORY SUPPLEMENT
(3.59-11)
M6 Adds space motion between e and J2000.0 to the position vector at e:
M6 =
I -tit
O I
(3.59-12)
where t\ = c(JD(e) - JD(J2000, 0)) / Q and c = tt / (180 x 3600).
The following two sections give complete algorithms for the conversion to and
from J2000.0 using the matrix M and its inverse M-1. It should be noted that here,
and in the method given on pages B42 and B43 of The Astronomical Almanac,
Kinoshita's development of Andoyer's precession angles (Aoki et al, 1983) have
been used. However, the numerical values for the vector components for correcting
the E-terms of aberration in Equation 3.591-2 have been taken from Yallop et al.
(1989).
Planetary positions may also be transformed using Mp, where Mp is a 3 x 3
matrix formed from partitioning M = M2M3M4M5 (M4 has no effect on position)
into four 3x3 matrices and extracting the leading matrix. Alternatively, use the
appropriate 3x3 rotational matrices from M2, M3, and M5.
3.591 Conversion of Stellar Positions and Proper Motions from the FK4 System
at B 1950.0 to FK5 System at J2000.0 A matrix method for calculating the mean
place of a star at J2000.0 on the FK5 system from the mean place at B1950.0 on
the FK4 system, ignoring the systematic corrections FK5-FK4 and individual star
corrections to the FK5, is as follows:
Step a: From a star catalog obtain the FK4 position (a0, 6q), in degrees, proper
motions {fiaa, fi$0) in seconds of arc per tropical century, parallax (ttq) in seconds of
arc, and radial velocity (v0) in kms"1 for B1950.0. If w0 or vq are unspecified, set
them both equal to zero.
Step b: Calculate the rectangular components of the position vector r0 and veloc-
ity vector r0 from
To =
ro =
COS Qo COS 60 1
sin cto cos <5o
sin60
— fiao sin Qo cos 60 — us,, cos qo sin So
+paf> cos a0 cos 60 — ^6o sin qo sin 60
fxs„ cos 60
(3.591-1)
+ 21.095 Vo7r0r0.
3 / CELESTIAL REFERENCE SYSTEMS 1 85
Step c: Remove the effects of the E-terms of aberration to form Ty and rj from
rL = r0 - A + (r0 ■ A)r0,
r, = r0 - A + (r0 ■ A)r0,
where A ~
A =
-1.62557
-0.31919
-0. 13843
+1'.'245
-1'.'580
-0"659
x 10 6 radians,
x 10 3 per tropical cy,
(3.591-2)
and (r0 • A) is the scalar product.
Step d: Form the vector Ri
and calculate the vector R :
from
R = MRi
(3.591-3)
where M is a constant 6x6 matrix:
+0.9999256782 -0.01 1 18206! 1 -0.0048579477 +0.OOOOO242395018 -0.00000002710663 - 0.00000001 177656
+0.0111820610 +0.9999374784 -0.0000271765 +0.00000002710663 +0.00000242397878 - 0.OOO0OOOOO06587
+0.0048579479 -0.0000271474 +0.9999881997 +0.00000001177656 -O.OOO0000O0O6582 +0.OO000242410173
-0.000551 -0.238565 +0.435739 +0.99994704 -0.01118251 -0.00485767
+0.238514 -0.002667 - 0.008541 +0.01118251 +0.99995883 - 0.00002718
-0.435623 +0.012254 +0.002117 +0.OO485767 - 0.00002714 +1.00000956
(3.591-4)
and set (x,y,z,x,y,z) = R'-
Step e: Calculate the FK5 mean position {au ffj), proper motions (jia,,fi6x) in
seconds of arc per Julian century, parallax (7Ti) in seconds of arc, and radial velocity
(vi) in kms-1 for .12000.0 from
cosqi cosi5] =xlr, sin at cos<5[ = ylr, sin<5i=z/r,
xy — yx
Mai — -j , 3" >
x1 + y1
zix2 + y1) - z(xx + yy)
vi = (xx + yy + zz) / (2 1 . 0957T0r), tt, = tt0 / r,
(3.591-5)
where r = y'jc2 + y2 + z2. If 7r0 is zero then v\ - vq-
1 fifi
EXPLANATORY SUPPLEMENT
3.592 Conversion of Stellar Positions and Proper Motions from the FK5 System
at J2000.0 to FK4 System at B1950.0 A matrix method for calculating the mean
place of a star at B1950.0 on the FK4 system from the mean place at J2000.0 on
the FK5 system, ignoring the systematic corrections FK4-FK5 and individual star
corrections to the FK4, is as follows:
Step 1: From a star catalog obtain the FK5 position (a„, Sq), in degrees, proper
motions (fiao,fiSo) in seconds of arc per Julian century, parallax (n0) in seconds of
arc, and radial velocity v0 in kms"1 for J2000.0. If tt0 or (v0) are unspecified, set
them both equal to zero.
Step 2: Calculate the rectangular components of the position vector r0 and veloc-
ity vector r0 from Equation 3.591-1.
Step 3: Form the vector Ro =
To
and calculate the vector R[ =
from:
R| =M-'Ro,
where M~' is a constant 6x6 matrix:
(3.592-1)
+0.9999256795 +0.0111814828 +0.0048590039 -0.00000242389840 -0.00000002710544 -0.00000001177742
-0.0111814828 +0.9999374849 -0.0000271771 +0.00000002710544 -0.00000242392702 +O0OOO0OOOOO6585
-0.0048590O40 -0.000027.557 +0.9999881946 +0.00O0000 1 177742 +O.000O000OO06585 -0 00000242404995
-0.000551 +0.238509 -0.435614 +0.99990432 +0.01118145 +0.00485852
-0.238560 - 0.002667 +0.012254 - 0.01118145 +0.99991613 -0 00002717
+0.435730 -0.008541 ,0.002,17 -0.00485852 -0.00002716 +0.99996684
(3.592-2)
Step 4: Include the effects of the E-terms of aberration as follows. Form s, = r, /r,
and S] = f i / n where r, - ^/x] +y~ + zj.
Set s = s, , and calculate r from r = S| + A - (s • A)s, where A is given in Step c of
Section 3.591.
Set s = rlr and iterate the expression for r once or twice until a consistent value of
r is obtained, then calculate
r = si + A - (s • A)s,
(3.592-3)
where A is given in Step c of Section 3.591.
3 / CELESTIAL REFERENCE SYSTEMS
187
Step 5: Calculate the FK4 mean (at}, 8\), proper motions (fiai,/iSl) in seconds of
arc per tropical century, parallax (tti) in seconds of arc, and radial velocity (v[) in
kms-1 for B1950.0, as given in Section 3.591 Step e, by setting (x,y,z) = r', (x, y, z) -
r' and r = ^Jx2 + y2 + z2.
In Step 4 set (xuyuzi) = r7,, (xuy\,Zi) = r1], and rj = <Jx\+y\ + z], then
V] =(xixt +y,yi +Zlj,)/(21.0957r0ri), tt, = tt0 / r] . (3.592^+)
If ttq is zero then v\ - v0.
3.6 REFERENCES
Those publications referenced in Chapter 3 are indicated by a |.
fAbalakin, V., ed. (1985). Refraction Tables of Pulkova Observatory, 5th edition, Cen-
tral Astronomical Observatory, Academy of Science of the USSR. Nauka Publishing
House, Leningrad Section, Leningrad, USSR.
f Andoyer, M.H. (1911). "Les Formules de la Precession d'apres S. Newcomb" Bull. Astron.
28, 67.
f Aoki, S., Guinot, B., Kaplan, G.H., Kinoshita, H., McCarthy, D.D., and Seidelmann, P.K.
(1982). "The New Definition of Universal Time" Astron. Astrophys. 105, 359.
Aoki, S. and Kinoshita, H. (1983). "Note on the Relation Between the Equinox and
Guinot's Non-rotating Origin" Celes. Mech. 29, 335.
f Aoki, S., Soma, M., Kinoshita, H., and Inoue, K. (1983). "Conversion Matrix of Epoch
B1950.0 FK4-based Positions of Stars to Epoch J2000.0 Positions in Accordance with
the New IAU Resolutions" Astron. Astrophys. 128, 263.
Aoki, S. (1988). "Relation Between the Celestial Reference System and the Terrestrial
Reference System of a Rigid Earth" Celes. Mech. 42, 309.
Arias, E.F., Lestrade, L.-F., and Feissel, M. (1988). "Relative Orientation of VLBI Celestial
Reference Frames" in The Earth's Rotation and Reference Frames for Geodesy and
Geodynamics A.K. Babcock and G.A. Wilkins, eds. IAU Symposium 128 (Kluwer,
Dordrecht, Holland), p. 61.
Ashby, N. and Allan, D.W. (1979). "Practical Implications of Relativity for a Global
Coordinate Time Scale" Radio Sci. 14, 649.
t Atkinson, R.d'E. (1972). "Modern Aberration for Observations of any Date" Astron. J.
77, 512.
t Atkinson, R.d'E. (1973). "On the 'Dynamical Variations' of Latitude and Time" Astron.
J. 78, 147.
t Atkinson, R.d'E. (1975). "On the Earth's Axes of Rotation and Figure" Mon. Not. R.
Astr. Soc. 171, 381.
Atkinson, R.d'E. (1976). "On the Earth's Axes of Rotation and Figure" Presented at XVI
Gen. Assembly of IUGG, Grenoble,
t Atkinson, R.d'E. and Sadler, D.H. (1951). "On the Use of Mean Sidereal Time" Mon. Not
R. Astr. Soc. Ill, 619.
188
EXPLANATORY SUPPLEMENT
f Auer, L.H. and Standish, E.M. Jr. (1979). "Astronomical Refraction; Computational
Method for all Zenith Angles" Yale University Astronomy Department.
Backer, D.C., Fomalont, E.B., Goss, W.M, Taylor, J.H., and Weisberg, J.M. (1985). "Ac-
curate Timing and Interferometer Positions for the Millisecond Pulsar 1937+21 and
the Binary Pulsar 1913+16" Astron. J. 90, 2275.
Barbieri, C. and Bernacca, P.L., eds. (1979). European Satellite Astrometry (Inst, di As-
tronomia, Univ. di Padova, Italy).
Bender, P.L. (1974). "Reference Coordinate System Requirements for Geophysics" in On
Reference Coordinate Systems for Earth Dynamics B. Kolaczek and G. Weiffenbach
eds. IAU Colloquium 26, (Smithsonian Astrophys. Obs, Cambridge, MA), p 85
Bender, P.L. and Goad, C. (1979). The Use of Artificial Satellites for Geodesy and Geo-
dynamics Vol. II (National Technical University, Athens).
tBlackwell, K.C. (1977). "Equinox— position and motion during 250 years" Mon. Not R
Astr. Soc. 180, Short Communication, 65P.
fBoss, B. (1937). General Catalog of 33S4S Stars for the Epoch 1950 Publication No. 468
(Carnegie Inst, of Washington).
Boucher, C. (1985). "Terrestrial and Inertial Relativists Reference Frames" Groupe de
Recherce en Geodesie Spatiale (GRGS) Rep. No. 3 (Institut de Geographie National
Paris).
Boucher, C. (1987). "Definition and Realization of Terrestrial Reference Systems for Mon-
itoring Earth Rotation" Institut de Geographic National/Service de la Geodesie et
du Nivellement (IGN/SGN), Paris No. 27, 459.
Boucher, C. and Altaminmi, Z. (1987). "Intercomparison of VLBI, LLR, SLR, and GPS
Derived Baselines on a Global Basis" Institut de Geographic National/ Service de la
Geodesie et du Nivellement (IGN/SGN), Paris No. 2, 450.
t Bretagnon, P. (1982) . "Theorie du mouvement de l'ensemble des planetes. Solution VSOP82'
Astron. Astrophys. 114, 278.
Brouwer, D. and de Sitter, W. (1938). "On the System of Astronomical Constants" Bull
Astron. Inst of the Netherlands 8, 213.
BIH Annual Rep. for 1984 (1985). (Bureau International de l'Heure, Paris)
BIH Annual Rep. for 1985 (1986). (Bureau International de l'Heure, Paris)
BIH Annual Rep. for 1986 (1987). (Bureau International de l'Heure, Paris)
t BIH Annual Rep. for 1987 (1988). (Bureau International de FHeure, Paris)
Cannon, W.H. (1978). "The Classical Analysis of the Response of a Long Baseline Radio
Interferometer" Geophys. J. R. Astr. Soc. 53, 503.
Capitaine, N, Williams, J.G., and Seidelmann, P.K. (1985). "Clarifications Concerning
the Definition and Determination of the Celestial Ephemera Pole" Astron. Astrophys.
-L--±U, Ool.
Capitaine, N. (1986). "The Earth Rotation Parameter: Conceptual and Conventional Def-
initions' Astron. Astrophys. 162, 323.
Capitaine, N Guinot, B., and Souchay, J. (1986). "A Non-Rotating Origm on the Instan-
taneous Equator: Definitions, Properties and Use" Celes. Mech. 39 283
Capitame, N. and Guinot, B. (1988). "A Non-Rotating Origin on the Instantaneous Equa-
a°v d Eartk'S Rotation and Reference Frames for Geodesy and Geodynamics
A.K Babcock and G.A. Wilkins, eds. IAU Symposium 128 (Kluwer, Dordrecht Hol-
land), p. 33.
Carter, W.E., Robertson, D.S., and Fallon, F.W. (1987). "Polar Motion and UT1 Time
Series Derived from VLBI Observations" BIH Annual Rep. for 1986 D-19, Paris.
3 / CELESTIAL REFERENCE SYSTEMS
189
Carter, W.E. (1988). "Corrections to the IAU 1980 Nutation Series" BIH Annual Rep. for
i£)£7D-105, Paris.
Chappie, W.M. and Tullis, T.E. (1977). "Evaluation of the Forces that Drive the Plates"
J. Geophys. Res. 82, No. 14, 1967.
Chauvenet, W., (5th ed., revised and corrected 1908), A Manual of Spherical and Practi-
cal Astronomy 2 vol. (Lippincott, Philadelphia), reprinted (Dover Publications, New
York, 1960).
Corbin, T.E. (1978). "The Proper Motions of the AGK3 R and SRS Stars" in Modern
Astrometry F.V. Prochazka and R.H. Tucker, eds. IAU Colloquium 48, (University
Observatory, Vienna), p. 505.
de Sitter, W. (1910). "On the Formula for the Comparison of Observed Phenomena of
Jupiter's Satellites with Theory" Mon. Not. R. Astr. Soc. 71, 85.
de Sitter, W. (1911). "On the Harvard Eclipses of Jupiter's Satellite IV" Mon. Not. R.
Astr. Soc. 71, 596 and 602.
Dicke, R.H. (1969). "Average Acceleration of the Earth's Rotation and the Viscosity of
the Deep Mantle" J. Geophys. Res. 74, 25, 5895.
Dickey, J.O. (1989). "Intercomparisons Between Kinematic and Dynamical Systems" in
Reference Frames in Astronomy and Geophysics J. Kovalevsky, I.I. Mueller, and B.
Kolaczek, eds. (Kluwer, Dordrecht, Holland), p. 305.
Duncombe, R.L., Seidelmann, P.K., and Van Flandern, T.C. (1974). "Celestial Reference
Systems Derivable from Solar System Dynamics" in On Reference Coordinate Sys-
tems for Earth Dynamics B. Kolaczek and G. Weiffenbach, eds. IAU Colloquium 26
(Smithsonian Astrophys. Obs., Cambridge, MA), p. 223.
Duncombe, R.L., Fricke, W., Seidelmann, P.K,, and Wilkins, G.A. (1976). "Joint Report of
the Working Groups of the IAU Commission 4 on Precession, Planetary Ephemerides,
Units and Time-Scales" Trans. IAU XVIB, 56.
t Dyson, Sir F.W., Eddington, A.S., and Davidson, C. (1920). "A Determination of the
Deflection of Light by the Sun's Gravitational Field, from Observations made at the
Total Solar Eclipse of May 29, 1919" Phil. Trans. R. Soc. London A220, 291.
Eichhorn, H.K. and Leacock, R.J. (1986). Astrometric Techniques IAU Symposium 109
(Reidel, Dordrecht, Holland),
t Einstein, A. (1956). The Meaning of Relativity (Princeton University Press, Princeton,
NJ).
t Emerson, B. (1973). "A Method of Calculating Apparent Places of Stars" R. Ohs. Bull.
178, 299.
Eubanks, T.M., Steppe, J.A., and Speith, M.A. (1985). "Earth Orientation Parameters
by Very Long Base Interferometry" BIH Annual Rep. for 1984 D"19. Paris.
t Explanatory Supplement to the Astronomical Ephemeris and the American Ephemeris and
Nautical Almanac (1961). (Her Majesty's Stationery Office, London, reprinted 1974).
fFabri, E. (1980). "Advocating the Use of Vector-Matrix Notation in Precession Theory"
Astron. Astrophys. 82, 123.
Fanselow, J.L., Sovers, O.J., Thomas, J.B., Purcell, G.H. Jr., Cohen, E.J., Rogstad, D.H.,
Skjerve, L.J., and Spitzmesser, D.J. (1984). "Radio Interferometric Determination of
Source Positions Utilizing Deep Space Network Antennas— 1971 to 1980" Astron. I.
89, 987.
Fedorov, E.P., Smith, M.L., and Bender, P.L., eds. (1980). Nutation and the Earth's Ro-
tation IAU Symposium 78 (Reidel, Dordrecht, Holland).
Feissel, M. (1980). "Determination of the Earth Rotation Parameters by the Bureau In-
ternational de l'Heure, 1962-1979" Bull. Geodesique 54, 81.
190
EXPLANATORY SUPPLEMENT
f FK3 (1937). "Dritter Fundamentalkatalog des Berliner Astromischen Jahrbuchs" Veroffent-
lichungen des Astronomischen Rechen-Instituts zu Berlin- Dahlem Nr. 54.
tFK4 (1963). "Fourth Fundamental Catalogue" Veroffentlichungen des Astronomischen
Rechen-Instituts Heidelberg No. 10.
tFomalont, E.B. and Sramek, R.A. (1975). "A Confirmation of Einstein's General Theory
of Relativity by Measuring the Bending of Microwave Radiation in the Gravitational
Field of the Sun" Astrophys. J. 199, 749.
Fricke, W. (1967). "Precession and Galactic Rotation Derived from Fundamental Proper
Motions of Distant Stars" Astron. J. 72, 1368,
Fricke, W. (1971). "A Rediscussion of Newcomb's Determination of Precession" Astron.
Astrophys. 13, 298.
Fricke, W. (1974) . "Definition of the Celestial Reference Coordinate System in Fundamen-
tal Catalogues" in On Reference Coordinate Systems for Earth Dynamics B. Kolaczek
and G. Weiffenbach, eds. IAU Colloquium 26 (Smithsonian Astrophys. Obs., Cam-
bridge, MA), p. 201.
f Fricke, W. (1977). "Basic Material for the Determination of Precession and of Galactic
Rotation and a Review of Methods and Results" Veroffentlichungen Astron, Rechen-
Inst. Heidelberg no. 28, Verlag G. Braun, Karlsruhe.
Fricke, W, and Gliese, W. (1978). "Progress in the Compilation of the FK5" in Modern
Astrometry F.V. Prochazka and R.H. Tucker, eds. IAU Colloquium 48, (University
Observatory Vienna), p. 421.
Fricke, W. (1979). "Progress Report on Preparation of FK5 for Commission 4" Proceedings
of the Seventeenth General Assembly" Montreal 1979, Trans. IAU XVIIB (Reidel,
Dordrecht, Holland), p. 65.
Fricke, W. (1979). "Methods of Compiling a Fundamental Reference System" in Collo-
quium on European Satellite Astrometry C. Barbieri and PL. Bernacca, eds. (1st. di
Astromomia, Univ. di Padova, Italy), p. 175,
t Fricke, W. (1982). "Determination of the Equinox and Equator of the FK5" Astron. As-
trophys. 107, L13.
Froeschle, M. and Kovalevsky J, (1982). "The Connection of a Catalogue of Stars with an
Extragalactic Reference Frame" Astron. Astrophys. 116, 89.
Gaposchkin, E.M. and Kolaczek, B. (1981). Reference Coordinate Systems for Earth Dy-
namics IAU Colloquium 56 (Reidel, Dordrecht, Holland).
Garfinkel, B. (1944). "An Investigation in the Theory of Astronomical Refraction" Astron.
J. 50, 169.
Garfinkel B. (1967). "Astronomical Refraction in a Polytropic Atmosphere" Astron. J. 72,
235.
t Gilbert, F. and Dziewonski, A.M. (1975). "An Application of Normal Mode Theory to the
Retrieval of Structural Parameters and Source Mechanisms from Seismic Spectra"
Phil. Trans. R. Soc. London A278, 187.
Goad, C. (1979). "Gravimetric Tidal Loading Computed from Integrated Green's Func-
tions" NOAA Tech. Memorandum NOS NGS 22 (NOS/NOAA, Rockville, MD).
Green, R.M. (1985). Spherical Astronomy (Cambridge University Press, Cambridge, Eng-
land),
Greenwich Observations for 1898 (1900). "Refraction Tables Arranged for Use at the
Royal Observatory, Greenwich" Appendix I, by P.H. Cowell (Her Majesty's Stationery
Office, London).
Guinot, B. (1978). Dept. of Geod. Sci. Rep. 280, LI. Mueller, ed. (Ohio State University,
Columbus), p. 13.
3 / CELESTIAL REFERENCE SYSTEMS
191
Guinot B (1979), "Basic Problems in the Kinematics of the Rotation of the Earth
trvneandthe Earth's Rotation D.D. McCarthy and J.D.H. Pilkmgton, eds. IAU
Symposium 82 (Reidel, Dordrecht, Holland), p. 7.
n ■ M R f 19791 "Irregularities of the Polar Motion" in Time and the Earth s Rotation
D D McCarthy and J D . H . PiUdngton, eds. IAU Symposium 82 (Reidel, Dordrecht,
Holland), p.,279. ^ ... -
Guinot B (1981). "Comments on the Terrestrial Pole of Reference, the Origin °f the Lon-
gitudes and on the Definition of UT1" in Reference Coordznate Systems for Earth
DynamicTEM. Gaposchkin and B. Kolaczek, eds. IAU Colloquium 56 (Re.del, Dor-
drecht, Holland), p. 125.
Guinot B (1986). "Concepts of Reference Systems" in Astrometnc TechmquesH.K E.ch-
horn and R J ■ Leacock, eds. IAU Symposium 109 (Reidel, Dordrecht, Holland), p. 1.
Guinot B (1989). "Astronomical and Atomic Time Scales" in Reference Frames in As-
Xonomy and Geophysics J. Kovalevsky, LI. Mueller and B. Kolaczek, eds. (Kluwer,
Dordrecht, Holland), p. 351.
Harzer P (1924). "Berechnung der Ablenkung der Lichtstrahlen in der Atmosphare der
Erde auf rein Meteorologisch-Physikalischer Grundlage" Publikation der Stemwarte
Harzer PM ^^''Gebrauchstabellen zur Berechnung der Ablenkungen der Lichtstrahlen
Tn der Atmosphare der Erde for die Beobachtungen am Grossen K.eler Mendiankre.se
Publikation der Sternwarte in Kiel No. 14.
Hellings, R.W. (1986). "Relativistic Effects in Astronomical Timing Measurements As-
tron J 91 651 (see also 92, 1446 for erratum).
Herring, T.A., Gwinn, C.R, and Shapiro, I.I. (1986). "Geodesy by Radio Interferometry:
Studies of the Forced Nutations of the Earth" J. Geophys. Res. 91, 47 45
t Herring, T.A. (1987). "Herrings Correction to the IAU 1980 Nutat.on Senes BIH Annual
^-/^TndTnclair AT (1985). "The Computation of Angular Atmospheric
Hohenkerk, C.Y., and binciair, a.i. li™;- . f not™ T^hnirnl Note 63.
Refraction at Large Zenith Angles" HM Nautical Almanac Office Technical Note M,
Royal Greenwich Observatory. ,
4.u i -i * | ,mW r>K M9821 "Results of Observations Made with the brx-mcn
»Hugh^:A™";«r;ai'6LB,„h.m,R.L, „« . »
Naval Observatory (in press). »mi A**embh
IAU (1948). "Commission 4, Ephemerides" Transactions of the Seventh General Assembly
Trans. IAU VII, 75, (Cambridge University Press, Cambridge, England)_
tIAU (1952). Transactions of the Eighth General Assembly Trans. IAU VIII, 67, JH, *
bridge University Press, Cambridge, England).
tIAU (1967). "Commission 4, Ephemerides" Preceding, of the Thirteenth General Assem
bly Trans. IAU XIIIB (Reidel, Dordrecht, Holland),
t IAU (1976). "Report of Joint Meetings of Commissions 4, 8, and 31 on the New System of
Astronomical Constants" Proceedings of the Sixteenth General Assembly Wff Trans.
IAU XVIB (Reidel, Dordrecht, Holland).
tIAU (1980). Reports on Astronomy Trans. IAU XVHIA (Reidel, Dordrecht, Holland), p. 8.
t Irwin, J.B. (1959). "Standard Light-Time Curves" Astron. J. 64, 149.
192
EXPLANATORY SUPPLEMENT
Jeffreys, W.H. (1980). "Astrometry with the Space Telescope" Celea. Mech. 22, 175.
Kaplan, G.H., ed. (1981). "The IAU Resolutions on Astronomical Constants, Time Scales,
and the Fundamental Reference Frame" United States Naval Observatory Circular 163
(U.S. Naval Observatory, Washington).
Kaplan, G.H., Josties, F.J., Angerhofer, P.E., Johnston, K.J., and Spencer, J.H. (1982).
"Precise Radio Source Positions from Interferometric Observations" Astron. J. 87,
570.
t Kaplan, G.H., Hughes, J. A., Seidelmann, P.K., Smith, C.A., and Yallop, B.D. (1989).
"Mean and Apparent Place Computations in the New IAU system. III. Apparent,
Topocentric, and Astrometric Places of Planets and Stars" Astron. J. 97, 1197.
Kaula, W.M. (1975). "Absolute Plate Motions by Boundry Velocity Minimizations" J.
Geophys. Res. 80, No. 2, 244.
fKinoshita, H. (1975). "Formulas for Precession" Smithsonian Astrophysical Observatory
Special Report 364, (Smithsonian Astrophys. Obs., Cambridge, MA).
fKinoshita, H. (1977). "Theory of the Rotation of the Rigid Earth" Celes. Mech. 15, 277.
Kinoshita, H., Nakajima, K., Kubo, Y., Nakagawa, L, Sasao, T., and Yokoyama, K. (1979).
"Note on Nutation in Ephemerides" Publ. Int. Lat. Obs. of Mizusawa XII, 71.
Kolaczek, B. and Weiffenbach, G., eds. (1974). "On Reference Coordinate Systems for
Earth Dynamics" IAU Colloquium 26, (Smithsonian Astrophys. Obs., Cambridge,
MA).
Kovalevsky, J. (1979). "The Reference Systems" in Time and the Earth's Rotation D.D.
McCarthy and J.D.H. Pilkington, eds. 82 (Reidel, Dordrecht, Holland), p. 151.
Kovalevsky, J. (1980). "Global Astrometry by Space Techniques" Celes. Mech. 22, 153.
Kovalevsky, J. and Mueller, LI. (1981). "Comments on Conventional Terrestrial and Quasi-
Inertial Reference Systems" in Reference Coordinate System for Earth DynamicsE.M.
Gaposchkin and B. Kolaczek, eds. IAU Colloquium 56 (Reidel, Dordrecht, Holland),
p. 375.
Kovalevsky, J. (1985). "Systemes de reference terrestres et celestes" Bull. Astronomique
de Observatoire Royal de Belgique 10, 87.
Kovalevsky, J. (1989). "Stellar Reference Frames" in Reference Frames in Astronomy and
Geophysics J. Kovalevsky, I.I. Mueller and B. Kolaczek, eds. (Kluwer, Dordrecht,
Holland), p. 15.
Kozai, Y. (1974). "Hybrid Systems for Use in the Dynamics of Artificial Satellites" in On
Reference Coordinate Systems for Earth Dynamics B. Kolaczek and G. Weiffenbach,
eds. IAU Colloquium 26 (Smithsonian Astrophys. Obs., Cambridge, MA), p. 235.
Larden, D. (1980). "Some Geophysical Effects on Geodetic Levelling Networks" Proceed-
ings of the Second International Symposium on Problems Related to the Redefinition
of North American Vertical Geodetic Networks (Canadian Inst, of Surveying, Ottawa).
tLederle, T. and Schwan, H. (1984). "Procedure for Computing the Apparent Places of
Fundamental Stars (APFS) from 1984 Onwards" Astron. Astrophys. 134, 1.
Leick, A. and Mueller, I.I. (1979). "Defining the Celestial Pole" Manus geodaetica 4, 149.
Lestrade, J.-F., Requieme, Y., Rapaport, M., and Preston, R.A. (1988). "Relationship
Between the JPL Radio Celestial Reference Frame and a Preliminary FK5 Frame"
in The Earth's Rotation and Reference Frames for Geodesy and Geodynamics A.K.
Babcock and G.A. Wilkins, eds. IAU Symposium 128 (Reidel, Dordrecht, Holland),
p. 67.
fLieske, J.H., Lederle, T., Fricke, W., and Morando, B. (1977). "Expressions for the Pre-
cession Quantities Based upon the IAU (1976) System of Astronomical Constants"
Astron. Astrophys. 58, 1.
3 / CELESTIAL REFERENCE SYSTEMS
193
fLieske, J.H. (1979). "Precession Matrix Based on IAU (1976) System of Astronomical
Constants" Astron. Astrophys. 73, 282.
Ma C. (1979). Very Long Baseline Interferometry Applied to Polar Motion, Relativity,
and Geodesy NASA Technical Memorandum 79582 (Goddard Space Flight Center,
Greenbelt, MD).
Ma, C. (1983). "Geodesy by Radio Interferometry: A Precise Celestial Reference Frame"
' EOS Transaction of the American Geophysical Union 64, 45, 674.
Ma, C, Clark, T.A., Ryan, J.W., Herring, T.A., Shapiro, I.I., Corey, B.E., Hinteregger,
H.F., Rogers, A.E.E., Whitney, A.R., Knight, C.A., Lundquist, G.L., Shaffer, D.B.,
Vandenburg, N.R., Pigg, J-C., Schupler, B.R. and Ronnang, B.O. (1986). "Radio-
source Positions from VLBI" Astron. J. 92, 1020.
Ma, C, Himwich, W., Mallama, A., and Kao, M. (1987). "Earth Orientation from VLBI"
' BIU Annual Rep. for 1986 D-ll, Paris.
Ma, C. (1988). "The VLBI Celestial Reference Frame of the NASA Crustal Dynamics
' Project" in The Earth's Rotation and Reference Frames for Geodesy and Geodynam-
ics A.K. Babcock and G.A. Wilkins, eds. IAU Symposium 128 (Reidel, Dordrecht,
Holland), p. 73.
Ma C. (1989). "Extragalactic Reference Frames" in Reference Frames in Astronomy and
Geophysics J. Kovalevsky, LI. Mueller and B. Koiaczek, eds. (Kluwer, Dordrecht,
Holland), p. 43.
f Markowitz, W. and Guinot, B., eds. (1968). Continental Drift, Secular Motion of the Pole
and Rotation of the Earth IAU Symposium 32 (Reidel, Dordrecht, Holland).
Markowitz. W. (1976). "Comparison of ILS, IPMS. BIH and Doppler Polar Motions with
Theoretical" Rep. to IAU Comm. 19 and 31, IAU Gen. Assembly, Grenoble.
Markowitz, W. (1979). "Independent Polar Motions, Optical and Doppler: Chandler Un-
certainties" Rep. to IAU Comm. 19 and 31, IAU Gen. Assembly, Montreal.
Mather, R.S., Coleman, R„ Rizos, C, and Hirsch, B. (1977). "The Role of Non-tidal
Gravity Variations in the Maintenance of Reference Systems for Secular Geodynam-
ics" UNISURV G 26, 1-25.
Mather, R.S., Masters, E.G., and Coleman, R. (1977). "A preliminary Analysis of GEOS-3
Altimeter Data in the Tasman and Coral Seas" UNISURV G 26, 27-46.
Mather, R.S. and Larden, D.R. (1978). "On the Recovery of Geodynamics Information
from Secular Gravity Change" UNISURV G 29, 11-22.
McCarthy, D.D. and Pilkington, J.D.H., eds. (1979). Time and the Earth's RotationlAU
Symposium 82 (Reidel, Dordrecht, Holland).
McClure, P. (1973). Diurnal Polar Motion GSFC Report X-592-73-259 (Goddard Space
Flight Center. Greenbelt, MD).
Melbourne, W., Anderle, R., Feissel, M., King, R., McCarthy, D.D., Smith, D.. Tapley, B.,
and Vicente, R. (1983). "Project MERIT standards" United States Naval Observatory
Circular 167 (U.S. Naval Observatory, Washington).
Melchior, P. and Yumi, S., eds. (1972). Rotation of the Earth IAU Symposium 48 (Reidel,
Dordrecht, Holland).
Melchior, P. (1978). The Tides of the Planet Earth (Pergamon Press, Oxford, England).
Minster, J.B. and Jordan, T.H. (1978). "Present-day Plate Motions" J. Geophys. Res. 83,
Bll, 5331.
Molodenskij, M.S. (1961). "The Theory of Nutation and Diurnal Earth Tides" Quatrieme
Symposium International sur les marees terrestres, Bruxelles, P. Melchior, ed. Com-
munications de l'Observatoire Royal de Belgique, No. 188, and Serie Geophysique,
No. 58, 25.
194
EXPLANATORY SUPPLEMENT
Morabito, D.D., Preston, R.A., Linfield, R.P., Slade, M.A., and Jauncey, D.L. (1986).
"Arcsecond Positions for Milliarcsecond VLBI Nuclei of Extragalactic Radio Sources:
IV, Seventeen Sources" Astron, J, 92, 546.
Moran, J.M. (1974). "Geodetic and Astrometric Results of Very Long-Baseline Interfero-
metric Measurements of Natural Radio Sources" in On Reference Coordinate Systems
for Earth Dynamics B. Kolaczek and G. Weiffenbach, eds. IAU Colloquium 26 (Smith-
sonian Astrophys. Obs., Cambridge, MA), pp. 269-292.
f Morgan, H.R. (1952). "Results of Observations with the Nine-Inch Transit Circle 1913—
1926" Publication of the United States Naval Observatory Second Series, XIII, Pt. III.
f Moritz, H. (1967). Dept. of Geod. Sci. Rep. 92, (Ohio State University, Columbus).
Moritz, H. (1979). Dept. of Geod. Sci. Rep. 294, (Ohio State University, Columbus).
tMoyer, T.D, (1981). "Transformation from Proper Time on Earth to Coordinated Time
in Solar System Barycentric Space-Time Frame of Reference (parts 1 and 2)" Celes.
Mech. 23, 33 and 57.
f Mueller, LI. (1969). Spherical and Practical Astronomy as Applied to Geodesy (Ungar,
New York).
Mueller, I.I. {1975}. "Tracking Station Positioning from Artificial Satellite Observations"
Geophys Surveys 2 243-276
Mueller, 1. 1., ed. (1975). Proceedings of the Geodesy/Solid Earth and Ocean Physics
(GEOP) Research Conferences 1972-1974. Dept. of Geod. Sci. Rep. 231 (Ohio State
University, Columbus).
Mueller, I.I., ed. (1978). "Applications of Geodesy and Geodynamics. An International
Symposium" Proceedings of the Geodesy/Solid Earth and Ocean Physics (GEOP)
Research Conference October 2-5, 1978, Columbus, OH Dept. of Geod. Sci. Rep. 280
(Ohio State University, Columbus).
Mueller, I.I. (1981). "Reference Coordinate System for Earth Dynamics: A Preview" in
Reference Coordinate System for Earth Dynamics E.M. Gaposchkin and B. Kolaczek,
eds. IAU Colloquium 56 (Reidel, Dordrecht, Holland), p. 1.
Mueller, I.I. (1985). "Reference Coordinate Systems and Frames: Concepts and Realiza-
tion" Bull. Geod. 59, 181.
Mueller, 1. 1., ed. (1985). Proceedings of the International Conference on Earth Rotation
and the Terrestrial Reference Frame (Dept. of Geod. Sci. and Surveying, Ohio State
University).
Mueller, I.I. and Wilkins, G.A. (1986). "Earth Rotation and the Terrestrial Reference
Frame" Adv. Space Res. 6, 9, 5.
Muhleman, D.O., Berge, G.L., Rudy, D.J., Niell, A.E., Linfield, R.P., and Standish, E.M.
Jr. (1985). "Precise Position Measurements of Jupiter, Saturn and Uranus Systems
with Very Large Array" Celes. Mech. 37, 329.
Murray, C.A. (1979). "The Epherneris Reference Frame for Astrometry" in Time and the
Earth's Rotation D.D. McCarthy and J.D.H. Pilkington, eds. 82 (Reidel, Dordrecht,
Holland), p. 165.
t Murray, C.A. (1981). "Relativistic Astrometry" Mon. Not. R. Astr. Soc. 195, 639.
I Murray, C.A. (1983). Vectorial Astrometry (Hilger, Bristol, England).
f Newcomb, S. (1895). The Elements of the Four Inner Planets and the Fundamental Con-
stants supplement to the American Epherneris and Nautical Almanac (USGPO, Wash-
ington, DC).
Newhall, XX, Preston, R.A., and Esposito, P.B. (1986). "Relating the JPL VLBI Reference
Frame and Planetary Ephemertdes" in Astrometric Techniques H.K. Eichhorn and
R.J. Leacock, eds. IAU Symposium 109 (Reidel, Dordrecht, Holland), p. 789.
3 / CELESTIAL REFERENCE SYSTEMS 1 95
Newhall, XX, Williams, J.G., and Dickey, J.O. (1987). "Earth Rotation (UTOR) from
Lunar Laser Ranging" BIH Annual Rep. for 1986 D-29, Paris.
Newton, I. (1968). "Experimental Evidence for a Secular Decrease in the Gravitational
Constant G" J. Geophys. Res. 73, 12, 3765.
Newton, R.R. (1974). "Coordinates Used in Range or Range-Rate Systems and Their Ex-
tension to a Dynamic Earth" in On Reference Coordinate Systems for Earth Dynamics
B. Kolaczek and G. Weiffenbach, eds. IAU Colloquium 26 (Smithsonian Astrophys.
Obs., Cambridge, MA), pp. 181-200.
| Paris Conference (1911). Congress International des Ephemerides Astronomiques tenu a
1'Observatoire de Paris du 23 au 26 October 1911.
Perley, R.A. (1982). "The Positions, Structures and Polarizations of 404 Compact Radio
Sources" Astron. J. 87, 859.
| Porter, J.G. and Sadler, D.H. (1953). "The Accurate Calculation of Apparent Places of
Stars" Mon. Not. R. Astr. Soc. 113, 455.
Prochazka, F.V. and Tucker, R.H., eds. (1978). Modern Astrometry IAU Colloquium, 48
(University Observatory, Vienna).
tPuIkova (1956). Refraction tables of Pulkova Observatory 4th edition, (Academy of Sci-
ences Press, Moscow, Leningrad).
Purcell, G.H. Jr., Cohen, E.J., Fanselow, J.L., Rogstad, D.H., Skjerve, L.J., Spitzmesser,
D.J., and Thomas, J.B. (1978). "Current Results and Developments in Astrometric
VLBI at the Jet Propulsion Laboratory" in Modern Astrometry F.V. Prochazka and
R.H. Tucker, eds. IAU Colloquium 48 (University Observatory, Vienna), p. 185.
Purcell, G.H., Jr., Fanselow, J.L., Thomas, J.B., Cohen, E.J., Rogstad, D.H., Sovers, O.J.,
Skjerve, L.J., and Spitzmesser, D.J. (1980). "VLBI Measurements of Radio Source
Positions at the Jet Propulsion Laboratory" Radio Interferometry Techniques for
Geodesy NASA Conference Publication 2115 (NASA Scientific and Tech. Information
Office, Washington, DC), p. 165.
Reasenberg, R.D. (1986). "Microsecond Astrometric Interferometry" in Astrometric Tech-
niques H.K. Eichhorn, and R.J. Leacock, eds. IAU Symposium 109 (Reidel, Dordrecht,
Holland), p. 321.
Reigber, Ch., Schwintzer, P., Mueller, H., and Massmann, F.H. (1987). "Earth Rotation
from Laser Ranging to LAGEOS: Station Coordinates from Laser Ranging to LA-
GEOS" BIH Annual Rep. for 1986 D-39, Paris.
Robertson, D.S., Fallon, F.W., and Carter, W.E. (1986). "Celestial Reference Coordi-
nate Systems: Submilliarcsecond Precision Demonstrated with VLBI Observations"
Astron. J. 91, 1456.
1 Saastamoinen J. (1972). "Introduction to Practical Computation of Astronomical Refrac-
tion" Bull. Geod. 106, 383.
tSAOC (1966). Smithsonian Astrophysical Observatory Star Catalog, Positions and Proper
Motions of 258997 Stars for the Epoch and Equinox of 1950. 0 Smithsonian Publication
4652 (Smithsonian Institution, Washington, DC).
Schutz, B.E., Tapley, B.D.. and Eanes, R.J. (1987). "Earth Rotation from LAGEOS Laser
Ranging" BIH Annual Rep. for 1986 D-33, Paris.
Schwan, H. (1986). "A New Technique for the Analytical Determination of a Fundamental
System of Positions and Proper Motions" in Astrometric Techniques H.K. Eichhorn
and R.J. Leacock, eds. IAU Symposium 109 (Reidel, Dordrecht, Holland), p. 63.
Schwan, H. (1988). "Construction of the System of Positions and Proper Motions in the
FK5" in Mapping the Sky S. Debarbat, J. A. Eddy, H.K. Eichhorn, and A.R. Upgren,
eds. IAU Symposium 133 (Kluwer, Dordrecht, Holland), p. 151.
196
EXPLANATORY SUPPLEMENT
Schwiderski, E. (1978). "Global Ocean Tides, Part 1; A Detailed Hydrodynamical In-
terpolation Model, US Naval Surface Weapons Center TR-3866 (U.S. Naval Surface
Weapons Center, Dahlgren, Virginia).
Scott, F.P. and Hughes, J.A. (1964). "Computation of Apparent Places for the Southern
Reference Star Program" Astron. J. 69, 368.
t Scott, F.P. (1964). "A Method for Evaluating the Elliptic E Terms of the Aberration"
Astron. J. 69, 372.
Seidelmann, P.K. (1982). "1980 IAU Theory of Nutation: The Final Report of the IAU
Working Group on Nutation" Celes. Mech. 27, 79.
f Shapiro, I.I. (1967). "New Method for the Detection of Light Deflection by Solar Gravity"
Science 157, 806.
Smart, W.M. (1944). Text-book in Spherical Astronomy 4th edition (Cambridge University
Press, Cambridge, England).
Smith, C.A. (1986). "On Desiderata for Star Catalogs for the Remainder of the Twentieth
Century: A Report on Catalog Work Now in Progress at the U.S. Naval Observatory"
in Astrometric Techniques H.K. Eichhorn and R.J. Leacock, eds. IAU Symposium 109
(Reidel, Dordrecht, Holland), p. 669.
f Smith, C.A., Kaplan, G.H., Hughes, J. A., Seidelmann, P.K., Yallop, B.D., and Hohenkerk.
C.Y. (1988). "Mean and Apparent Place Computations in the New IAU System. I.
The Transformation of Astrometric Catalog Systems to the Equinox J2000.0" Astron.
J. 97, 265.
Solomon, S.C. and Sleep, N.H. (1974). "Some Simple Physical Models for Absolute Plate
Motions" J. Geophys. Res. 79, 17, 2557.
Sovers, O.J., Edwards, C, Jacobs, C, Lanyi, G., Liewer, K., and Treuhaft, R. (1989).
"Astrometric Results of 1978-1985 Deep Space Network Radio Interferometry: The
JPL 1987-1 Extragalactic Source Catalog" Astron. J. 95, 1647.
tStandish, E.M. Jr. (1982a). "Orientation of the JPL Ephemerides, DE200/LE200, to the
Dynamical Equinox of J2000.0" Astron. Astrophys. 114, 297.
tStandish, E.M. Jr. (1982b). "Conversion of Positions and Prop er Motions from B1950.0
to IAU System at J2000.0" Astron. Astrophys. 115, 20.
Stolz, A. and Larden, D.R. (1979). "Seasonal Displacement and Deformation of the Earth
by the Atmosphere" J. Geophys. Res. 84, Bll, 6185.
fStumpff, P. (1979). "The Rigorous Treatment, of Stellar Aberration and Doppler Shift,
and the Barycentric Motion of the Earth" Astron. Astrophys. 78, 229.
fStumpff, P. (1980a). "On the Relationship between Classical and Relativistic Theory of
Stellar Aberration" Astron. Astrophys. 84, 257.
fStumpff, P. (1980b). "Two Self-Consistent Fortran Subroutines for the Computation of
the Earth's Motion" Astron. Astrophys. Supp. Ser. 41, 1.
fStumpff, P. (1985). "Rigorous Treatment of the Heliocentric Motion of Stars" Astron.
Astrophys. 144, 232.
Stumpff, P. (1986). "Relativistic and Perspective Effects in Proper Motions and Radial
Velocities of Stars" in Relativity in Celestial Mechanics and Astronomy J. Kovalesky
and V.A. Brumberg, eds. IAU Symposium 114, (Reidel, Dordrecht, Holland), p. 193.
f Taff, L.G. (1981). Computational Spherical Astronomy (Wiley, New York),
t The Astronomical Almanac (1984). (U.S. Government Printing Office, Washington, and
HM Stationery Office, London).
Van Altena, W. (1978). "Space Telescope Astrometry" in Modern Astrometry F.V. Proc-
hazka and R.H. Tucker, eds. IAU Colloquium 48, (University Observatory, Vienna),
p. 561.
3 / CELESTIAL REFERENCE SYSTEMS
197
Van Flandern T C (1981). "The Introduction of the Improved IAU System of Astronom-
ical Constants Time Scales and Reference Frame into The Astronomical Almanac"
The Supplement to the Astronomical Almanac 1984 (U.S. Government Printing Office,
Washington, and HM Stationery Office, London), p. S26.
Van Hylckama, T.E.A. (1956). "The Water Balance of the Earth" Drexel Institute of
Technology, Publications in Climatology 9, 59-117.
Vanicek, P. (1980). "Tidal Corrections to Geodetic Quantities" NOAA Tech. Rep. NOS
83 NGS 14 (NOS/NOAA, Rockville, MD).
Vondrak J. (1983). "On the Indirect Influence of the Planets on Nutation. I. Effects of
Planetary Perturbations in Lunar Orbit" Bull. Astron. Inst. Czechosl. 34, 184.
t Vondrak J. (1983). "On the Indirect Influence of the Planets on Nutation. II. Effects of
Planetary Perturbations of the Earth's Orbit" Bull. Astron. Inst. Czechosl. 34, 311.
Wade C M and Johnston, K.J. (1977). "Precise Positions of Radio Sources. V. Positions
of 36 Sources on a Baseline of 35 km" Astron. J. 82, 791. Erratum Astron. 7 84, 1932.
Wahr J.M. (1979). "The Tidal Motions of a Rotating, Elliptical, Elastic and Oceanlcss
Earth" Ph.D. thesis (Dept. of Physics, University of Colorado, Boulder),
f Wahr, J.M. (1981). "The Forced Nutations of an Ellipitical, Rotating, Elastic, and Ocean-
less Earth" Geophys. J. R. Astr. Soc. 64, 705.
Washington (1904). Tables XXVI XXXIV of "Reduction Tables for Transit Circle Ob-
servations" Publications of the United States Naval Observatory, Second Series, 4,
Wilkti^G.A^nd Mueller, LI. (1986). "Joint Summary Report of the IAU/IUGG Working
Groups on the Rotation of the Earth and Terrestrial Reference System in Highlights
of Astronomy J.P. Swings IAU Symposium 7 (Reidel, Dordrecht, Holland), p. 771
Williams J.G. and Melbourne, W.G. (1982). "Comments on the Effect of Adopting New
Precession and Equinox Corrections" in High-precision Earth Rotation and Earth-
Moon Dynamics O. Calame, ed. IAU Colloquium 63 (94), (Reidel, Dordrecht, Hol-
land), p. 293. , „, .c
Williams, J.G., Dickey, J.O., Melbourne, W.G., and Standish, E.M. Jr. (1983). Unifica-
tion of Celestial and Terrestrial Coordinate Systems" Proceedings of IAG Symposia,
IUGG XVIII General Assembly, Hamburg, 2 (Dept of Geod. Sci. and Surveying, Ohio
State University), p. 493.
Williams, J.G. and Standish, E.M. Jr. (1989). "Dynamical Reference Frames in the Plan-
etary and Earth-Moon Systems" in Reference Frames in Astronomy and Geophysics
J. Kovalevsky, I.I. Mueller and B. Kotaczek, eds. (Kluwer, Dordrecht, Holland , p. bt.
Willis, J.E. (1941). "A Determination of Astronomical Refraction from Physical Data
Transactions of the American Geophysical Union, part II, 324.
Winkler, G.M.R. and Van Flandern, T.C. (1977). "Ephemeris Time, Relativity, and the
Problems of Uniform Time in Astronomy" Astron. J. 82, 84.
fWoolard, E.W. and Clemence, G.M. (1966). Spherical Astronomy (Academic Press, New
York and London). nfR
t Yallop, B.D. (1984). in The Astronomical Almanac 1984 (U.S. Government Printing Omce,
Washington, and HM Stationery Office, London), pp. B16-B41.
t Yallop, B.D., Hohenkerk, C.Y., Smith, C.A., Kaplan, G.H., Hughes, J.A., and Seidelmann,
P.K. (1989). "Mean and Apparent Place Computations in the New IAU System.
II. Transformation of Mean Star Places from FK4 B1950.0 to FK5 J2000.0 Us.ng
Matrices in 6-space" Astron. J. 97, 274.
tYatskiv, Y.S. (1980). "Nearly Diurnal Free Polar Motion Derived from Astronomical Lat-
itude and Time Observations" in Nutation and the Earth's Rotation E.P. I-edorov,
198
EXPLANATORY SUPPLEMENT
M.L. Smith and P.L. Bender, eds. IAU Symposium 78 (Reidel, Dordrecht, Holland)
p. 59.
Zhu, S.Y. and Mueller, I.I. (1983). "Effects of Adopting New Precession, Nutation and
Equinox Corrections on the Terrestrial Reference Frame" Bull. Geodes. 57, 29.
Zverev, M.Z., Polozhentsev, D.D., Stepanova, E.E., Khrutskaya, E.V., Yagudin, L.I., and
Polozhentsev, A.D. (1986). "On the SRS Catalogue" in Astrometric Techniques H.K.
Eichhorn and R.J. Leacock, eds. IAU Symposium 109 (Reidel, Dordrecht, Holland)
p. 691.
CHAPTER
Terrestrial Coordinates
and the Rotation of the Earth
by Brent A. Archinal
4.1 INTRODUCTION
The task of establishing or defining the terrestrial coordinates of a point is inextrica-
bly linked to establishing the rotation of the Earth over time. The definition of any
terrestrial reference coordinate system (TRS) is given by establishing a celestial
reference coordinate system (CRS) and a suitable transformation between them.
This transformation is what is generally called Earth rotation (or Earth orienta-
tion), including the various changes in the orientation of the Earth's axis in space
and internally, and the rotation (and rotational variations) about that axis. (This
latter rotation is also often defined as Earth rotation — both uses of this term will
be made here with the meaning being given from context.) One could connect the
celestial and terrestrial coordinate system at any given instant using three Eulerian
angles, and then form a time series of such angles to define Earth rotation. However,
for historical reasons {mostly in order to associate the motions of the Earth with
physical processes and to allow for successive approximations of the motion), in
practice these angles are expressed in terms of precession, nutation, polar motion,
and rotation about the Earth's axis.
Before we proceed into any explanation of terrestrial coordinates, it is also nec-
essary to distinguish between a (celestial or terrestrial) "reference system" and a
"reference frame." A reference systern'm the conceptual definition of an "ideal' coor-
dinate system, based on some abstract principle(s). For example, an ideal terrestrial
system would be one in which the Earth's crust shows only deformation with no
rotations or translations. A conventional reference system is one where the model
used to define coordinates is given in detail; that is, specific model choices have
been made. The realization of such a conventional reference system, using actual
199
200 EXPLANATORY SUPPLEMENT
observations, adopted station coordinates, etc., provides a conventional reference
frame (Kovalevsky and Mueller, 1981; Mueller, 1988; Kovalevsky et al, 1989).
This chapter initially will provide the basic concepts of various adopted terres-
trial reference frames (mostly based on the use of reference ellipsoids), and of the
Conventional Terrestrial Reference Coordinate System (CTRS) as it is currently
realized by the International Earth Rotation Service (IERS). Also included is in-
formation on the actual transformation of coordinates from one frame to another.
This will be followed by information, mostly extracted from the IERS Standards
(McCarthy, 1990), regarding the Earth's gravity field (to a limited extent), tides,
and plate tectonics. Methods for monitoring the Earth's rotation (past, present, and
future) are described and information on the IERS is provided. Finally, a summary
of determinations of historical variations in the rotation of the Earth's axis is given.
An overview of Earth rotation and its causes in general has not been given
because space is lacking for such a complete treatment, and the ample coverage from
this viewpoint by Teisseyre (1989), Moritz and Mueller (1987), Lambeck (1980),
and Munk and MacDonald (1960). The motion of the Earth's celestial pole due to
precession and nutation has already been covered in Chapter 3. (Readers seeking
more detailed information on the subjects of gravity, geodesy, geodetic astronomy,
and terrestrial reference systems should consult standard texts such as Mueller and
Rockie, 1966; Heiskanen and Moritz, 1967; Mueller, 1969; Bomford, 1971; Moritz,
1980; Torge, 1980; Moritz and Mueller, 1987; Kovalevsky et al, 1989.) McCarthy
(1990) also provides further information on accepted models of effects on various
space geodetic systems.
4.2 TERRESTRIAL COORDINATE SYSTEMS
In this section the fundamentals of various terrestrial coordinate systems are pre-
sented, covering both the various worldwide and common local systems used to
define the coordinates of an object on or near the Earth's surface. It includes the
necessary explanations of the common abstract representations of the Earth's sur-
face, such as with the geoid or various ellipsoids. It ends with information on the
current Conventional Terrestrial Reference Coordinate System, as it is realized by
the IERS Terrestrial Coordinate System.
4.21 The Figure of the Earth and the Reference Ellipsoid
Most terrestrial coordinate systems for measuring positions near the Earth's surface
use as their reference an abstract surface near that of the Earth. A best-fitting
rotational ellipsoid is most commonly assumed as an approximation to the surface
of the Earth. A rotational ellipsoid is used, since it represents the shape of the
Earth far better than a sphere, because of the flattening of the Earth at the poles,
TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
201
Topography
Geop (through p)
Geoid
Ellipsoid
Figure 4.21.1
The Earth's surface and reference surfaces
and yet is still a fairly simple mathematical surface. (A more complicated surface,
such as a triaxial ellipsoid, will not greatly improve the representation, but will
greatly complicate the mathematics involved in using it.) Such a reference surface
has proven useful for referencing the position of points and measurements on or near
the Earth's surface. Reference ellipsoids may be either local or terrestrial, depending
on whether they are an approximation for a given area or the entire Earth.
Another type of hypothetical surface that proves useful for referencing obser-
vations is a geopotential surface, or geop, a surface of equal gravity potential. (In
geodesy, gravity includes both the effects of gravitational potential and potential
due to centrifugal force.) The gravity vector is perpendicular to the geop at every
point. Such a definition is quite natural when most instruments operated on the
surface of the Earth are sensitive to local gravity (through the use of a plumb bob,
bubble level, pool of Mercury, etc.). It is also critical for any hydrologic purpose,
where any water surface closely follows a geop, and where water flows "downhill"
from a geop of higher potential to one of lower potential. A particular geop of
great importance is the geoid, a geop whose potential is defined to be that of the
Earth at sea level. Many different definitions have been given in the past for the
geoid, perhaps the best conceptually being an imaginary "surface coinciding with
mean sea-level in the oceans, and lying under the land at the level to which the
sea would reach if admitted by small frictionless channels" (Bomford, 1971). Vari-
ous other definitions are given in Heiskanen and Moritz (1967, pp. 48-50); Moritz
(1980, p. 6); and Torge (1980, pp. 2-3). In practice, the average gravity potential
of the Earth at sea level is determined empirically (nowadays usually via altimet-
ric measurements of the world's oceans by satellite), and the surface of the same
potential is by definition the geoid. The geoid surface is then effectively the same
surface usually referred to as mean sea level.
The over all situation is shown in Figure 4.21.1. The observer is at point p
on the surface of the Earth (topography). A reference ellipsoid (particularly a local
one) approximates the geoid and lies generally below the topography. A geop passes
through p, but the geoid itself lies close to the reference ellipsoid. Note that although
202
EXPLANATORY SUPPLEMENT
Figure 4.22.1
Geocentric and geodetic latitude
the geop and geoid surfaces are not the same as that of the topography, there are
some similarities since the mass of the topography influences the local gravity. Also
shown in the figure are the geoid undulation (A/), which is the height of the geoid
above the ellipsoid, the ellipsoidal height (h), which is the height of p above the
ellipsoid, and the mean sea level height or orthometric height (H), which is the
height of p above the geoid. Note that:
N~h — H. (4.21-1)
This relationship is approximate since H is actually measured along the curved
vertical or plumb line through p, whereas h and N are measured along a straight,
line through p perpendicular to the ellipsoid.
4.22 Geocentric, Geodetic, and Astronomical Coordinates
Various coordinate systems exist for measuring longitude and latitude. Geocen-
tric coordinates are determined from the center of the Earth; geodetic coordinates
are determined relative to a reference ellipsoid; and astronomical (or geographic)
coordinates are determined relative to the observer's local vertical. Figure 4.22.1
illustrates the difference between geocentric and geodetic latitude, once again for
a point p on the Earth's surface. The geocentric latitude <j>' is the angle between
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
203
z
Figure 4.22.2
Astronomic latitude and longitude
the ellipsoidal equator and line from p to the center of the ellipsoid. The geodetic
latitude 0 is the angle between the ellipsoidal equator and the normal from p to
the ellipsoid (at point p')\ a and b are the semi-major and semi-minor axes of the
ellipsoid, respectively; /, the flattening, or 1 //, the inverse flattening, is often used,
with
a
The geocentric longitude (A) is defined by the angle between the reference (or zero)
meridian and the meridian of point p (and p'), measured eastward around the Earth
from 0° to 360° (IAU, 1983, p. 47). The geodetic longitude will be the same as the
geocentric longitude if the reference ellipsoid has been chosen with the same axes
and reference meridian as the Earth — this is assumed to be the case in further
discussion here. The astronomical coordinates of point p are determined relative to
the local vertical direction (or zenith) at point p (see Figure 4.22.2, where z indicates
the direction of the zenith and z' the direction of the ellipsoidal normal). A vertical
(or plumb) line passing through p will intersect the Earth's equatorial plane with
the angle the astronomical latitude of p. Astronomical longitude A is measured
as the angle between the reference meridian plane and the plane including the local
vertical through p and perpendicular to the equatorial plane. Note that the local
204
EXPLANATORY SUPPLEMENT
Figure 4,22.3
Deflection of the vertical on a unit sphere
vertical does not necessarily pass through the Earth's axis. The angle between the
local vertical direction and ellipsoidal normal through p is known as the deflection
of the vertical 9. The deflection of the vertical is often broken into two orthogonal
components (see Figure 4.22.3, where z indicates the direction of the zenith, and zf
the direction of the ellipsoidal normal), one £ being in the meridional direction, the
other Tj in the orthogonal prime vertical direction. Relations between the geodetic
and astronomical longitude and latitude and these components of the deflection of
the vertical are
sin <f> = cos ?; sin($ - f), (4.22-2)
sin t? = cos 4> sin( A - A). (4.22-3)
Using small-angle formulas, these can also be given as
£ = (4.22-4)
7? = (A- A)cos0. (4.22-5)
As to the measurement of the "height" of point p, several types of "height"
measurements are possible, referred either to a nearby surface, such as the geoid
or ellipsoid, or to the center of the Earth. The ellipsoidal height h, and mean sea
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
205
meridian/
e/
V
/' l \
Figure 4.22.4
Geocentric cartesian coordinates
level or geoidal height H have already been mentioned. The geocentric radius r,
is also shown in Figure 4.22.4. These heights are associated with the geodetic,
astronomical, and geocentric systems, respectively.
In addition to these systems of latitude, longitude, and height, it is common
to express the position of a point on or near the Earth in a right-handed geocen-
tric cartesian coordinate system (Figure 4.22.4). The coordinates of a point p are
expressed as a triplet including X, Y, and Z coordinates. Here the direction of the
Z-axis is that of the rotational reference ellipsoid, and the X-axis is perpendicular
to that through the reference meridian. The K-axis is of course perpendicular to
the X- and F-axes in a right-handed sense. As with the other systems, the origin is
(ideally) the center of the Earth. Conversion from geocentric to geocentric cartesian
coordinates is given by
X
Y
Z
J p
cos 4>' cos A
cos 4>' sin A
sin <f>'
(4.22-6)
206 EXPLANATORY SUPPLEMENT
Conversion from geodetic to geocentric cartesian coordinates is given by
X
Y
Z
(N$ + h) cos 4> cos A
(N^ + h) cos (j> sin A
((1 - e2)A^ + h) sin cf>
(4.22-7)
where e is the eccentricity of the ellipsoid
\/o2 - b2 ,
= V2f-f, (4.
and Nj, is the ellipsoidal radius of curvature in the meridian, and is given by
V 1 — e1 sin2 4>
Conversion from geocentric cartesian to geocentric coordinates is given by
" 4>' '
A
r
p
tan
- 1
tan
-i 1
z
(4.22-10)
s/x2 + y2+z2
Note that = sin 1 Z may also be used to determine <j>' , and that these functions
are evaluated such that -90° < <j> (or <p') < +90°, and 0° < A < 360°.
Conversion from geocentric cartesian to geodetic coordinates is not as simple
a matter, however. This problem has long been an interesting one for geodesy.
However, Borkowski (1989) has recently reviewed the various techniques used to
solve this problem, and has proposed two solutions, one of which is exact and is
reproduced here. It is based on using an expression for the reduced latitude in a
solvable fourth-degree polynomial.
Several intermediate values are computed in order to perform a solution:
r = s/x2 + Y2,
(4.22-11)
£_ [bZ-(a2-
ar
-b1)]
(4.22-12)
f_ [bZ+ia2-
ar
b2)]
J
(4.22-13)
P = -(£F+1),
(4.22-14)
Q = 2{E2 - F2),
(4.22-15)
D = P3 + Q2,
(4.22-16)
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
1
G =
t =
y/E2 + v + E
G2 +
F-vG
2G-E
207
(4.22-17)
(4.22-18)
(4.22-19)
Finally, the latitude and ellipsoidal height are computed from:
1 -f2"
0 ■ arctan
2bt J
h = (r - at) cos 4> + (z-b) sin <f>.
(4.22-20)
(4.22-21)
If D < 0, e.g., if less than about 45km from the Earth's center, the following
equation should be used for v in order to avoid the use of complex numbers:
v = 2\
4 - arccos
Q
(-py
(4.22-22)
To obtain the proper sign (and solution of the fourth-degree polynomial since up
to four solutions actually exist), the sign of b should be set to that of z before
beginning. Borkowski also notes that, of course, this solution is singular for points
at the z-axis (r = 0) or on the X7-plane (Z = 0). Additionally for points close to
those conditions, some roundoff error may be avoided and the accuracy improved
slightly by replacing the value of v with
v3 + 2Q
3P
Finally,
A = arctan
(4.22-23)
(4.22-24)
4.23 Local Coordinate Systems
Many possibilities exist for establishing local coordinate systems, either systems
designed for a specific observer or systems designed to cover a given local area. Most
systems specific to an observer are horizon (or altitude and azimuth) coordinate
systems and a "standard" one is described below. Coordinate systems designed to
cover a localized area are usually based on a "map projection" coordinate system,
such that the coordinates of a point relative to the reference ellipsoid are projected
onto a plane upon which computations can be done with relative simplicity (as
opposed to working on the ellipsoid itself). Some examples of the more commonly
208
EXPLANATORY SUPPLEMENT
used of these types of systems are given, such as for Universal Transverse Mercator
coordinates and (in the U.S.) state plane coordinates.
4.231 Horizon System Depending on the author, slight variations exist in the def-
inition of any local horizon coordinate system. However, a more or less "standard"
system is described as follows.
First, any such system must have at least two reference planes. The primary
reference plane here is the local horizon, and the secondary is the local meridian
(see Figure 4.231.1). The horizon is the plane perpendicular to the observer's local
vertical, while the meridian is the plane perpendicular to the horizon that passes
through the celestial poles. The direction of any point p2 (from the observer at
pi ) in this system is then given by its altitude and azimuth. If we imagine a plane
passing through p\ and the zenith z, the angle between the horizon and the point
p2 in this plane is the altitude (also sometimes known as the elevation angle) a,
and the angle between the meridian and this plane measured clockwise from north,
as viewed from above, is the azimuth A. (Some users in the past, particularly for
geodetic purposes, have defined azimuth as being measured clockwise from south.
This practice is no longer much in use but users of old observations should be aware
of it.) The altitude will have an angular value of from —90° to +90° (negative if
Figure 4.231.1
Altitude and azimuth
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 209
the object is below the horizon and positive if above), and the azimuth will have
an angular value of from 0° to 360°. Instead of this spherical coordinate system,
a left handed three-dimensional cartesian coordinate system is also used, with the
z-axis positive up, the x-axis positive north, and the y-axis positive east (through
90° azimuth).
A further variation of this system is to use the zenith distance z instead of
the altitude. The zenith distance is simply the complement of the altitude, i.e.,
the angular distance of p2 from the zenith z- It has the advantage of always being
positive, with angular values of from 0° to 180°. It is computed from altitude as
z = 90°-a, (4.231-1)
where the z and a are in degrees.
4.232 Grid Coordinate Systems For computational convenience, it is common to
represent sections of the curved surface of the Earth with a plane. Various systems
have been devised to do this, all based on given transformation equations from the
sphere or ellipsoid to the plane. These transformations are, in fact, map projections,
in many cases the same as those used for mapping even large segments of, or all of
the Earth. Conformal projections are usually used so that directions on the surface
being mapped are preserved on the grid. The coordinate systems based on such
map projections are usually termed grid coordinate systems. The most widely used
are the Universal Transverse Mercator, and the state plane coordinate systems of
the United States. Obviously many other systems exist for various countries and
areas, but the basic principles (and many times the projection formulas, with the
exception of constants) are the same.
For any grid coordinate system, several fundamental transformations and quan-
tities are of importance. The fundamental transformations are the direct conversion
from geodetic coordinates {<p, A) to grid coordinates {X, 30, and the inverse transfor-
mation from grid coordinates to geodetic coordinates. Other quantities of interest
include: (a) the convergence of the meridians, i.e., the angle between grid north
and true north on the grid (7); (b) the angle on the grid between the geodesic (the
shortest distance between two points on the ellipsoid) and the straight line between
them (known in some of the references as "r-T"); (c) the scale at a given point (k);
and (d) the scale factor along a line. Equations for the transformations and scale
are given below (based on Snyder (1983)); equations for the other quantities are
available from the other references, particularly Thomas (1952).
Fundamental quantities and expressions used are summarized as follows:
<p = geodetic latitude of point;
A = geodetic longitude of point;
<t>' = foot point latitude (see definition below);
210
EXPLANATORY SUPPLEMENT
A0 = longitude of the origin (and the central meridian for the transverse Mer-
cator grid);
<f>u <h = latitude of the standard parallels for Lambert conformal conic grid;
AA = difference of longitude from the central meridian (AA = A - A0) (in
radians);
a — semi-major axis of the ellipsoid;
h = semi-minor axis of the ellipsoid;
2 — b2
e = eccentricity of the ellipsoid (e2 = r — );
e -
3 iP
12 a " >
second eccentricity of the ellipsoid (e ~ =
b ■
a(l -e2)
= ellipsoidal radius of curvature in the meridian (Nj, - - _ ^_ ^ ^3/, ) .
Fortran programs for these transformations are available from the U. S. Na-
tional Geodetic Survey (NGS, 1985). For more information write National Oceanic
and Atmospheric Administration, National Geodetic Survey (N/CG174), Rockville,
MD 20852.
4.233 Universal Transverse Mercato rand Universal Polar Stereographic The Uni-
versal Transverse Mercator (UTM) [along with the Universal Polar Stereographic
(UPS)] coordinate system is a worldwide grid system of separate zones. It was de-
veloped in 1947 by the U. S. Army as a worldwide military grid system. Originally
designed to be useful for artillery operations, it has since been used commercially
as well. The UTM system consists of sixty 6° wide zones of longitude, with zone 1
going from -180° to -174° longitude, zone 2 from -174° to -168° longitude, etc.
Letter designations are sometimes applied to zones running from south to north.
The various zones are as shown in Figure 4.233.1. The central meridian is the cen-
tral meridian of each zone, with a scale factor of kQ = 0.9996. This results in a
maximum scale error of 1/1000 anywhere in a zone, or 1/2500 for points on the
central meridian or above 45° or below -45° latitude. (Of course, as with all grid
systems, there is no scale error if the appropriate corrections for scale are made.)
The use of UTM is usually limited to between +84° and -80° latitude. Grid coor-
dinates are defined with X = 500000 meters for all points on the central meridian,
and Y = 0 meters for all points on the equator when points are being referenced in
the Northern Hemisphere, and Y = 10000000 meters when points are being refer-
enced in the Southern Hemisphere. Although the latitude and longitude of a point
referenced to any ellipsoid may be converted to UTM coordinates, for certain areas
of the world, specific ellipsoids are associated with UTM coordinates. For example
the specified ellipsoids are Clarke 1866 for North America, Clarke 1880 for Africa,
and the International for most of the rest of the world, except for the far east and
Australia. [See Dept. of the Army (1973, plate 1) for these ellipsoid designations.
Section 4.24 explains the subject of ellipsoids in more detail.]
212 EXPLANATORY SUPPLEMENT
The Universal Polar Stereographic Grid System (UPS) consists of two polar
stereographic projections, one for the north polar area and one for the sooth polar
area. The north zone covers all areas north of 84° latitude, and the south zone all
areas south of -80° latitude. The scale factor at the origin (the poles) is k0 - 0. 994.
In both zones, a 2000000 meter ^-coordinate line coincides with the 0° and 180°
meridians. The pole will have coordinates pfX- 2000000 meters, and Y = 2000000
meters. The international ellipsoid (a = 6378388 meters, f = 11 297) is associated
with UPS coordinates. Because of the areas it covers, the UPS is little used. [For
more information see Dept. of the Army (1958).]
Although UTM and UPS were originally developed for use with tables and
graphs, with modern computers it is easier to use the actual transformation equa-
tions. The transformation procedures are given below [from Snyder (1983, pp. 67-
69, 160-164)]; both UTM and UPS are also discussed in Parker and Bartholomew
(1989).
For UTM, starting with our previously given expressions, we continue by com-
puting 5^, the length of the meridional arc, the true meridional distance on the
ellipsoid from the equator to latitude 0:
= a[{\ - e2 1 4 - 3e4 / 64 - 5e6 1256 - -)4>
- Qe2 ! 8 + 3e4 / 32 + 45<?6 / 1 024 + ■ • •) sin 20
+ (15e4 / 256 + 45/ / 1024 + •■■) sin 40
- (35e6 / 3072 + -) sin 60 + -], (4.233-1)
where 0 is in radians. Also needed are other auxiliary quantities:
T = tan2 0, (4.233-2)
C = e'2 cos2 0, (4.233-3)
A = cos 0(AA). (4.233^1)
We also compute as given by S computed at the latitude 0O at the origin of the
X and Y coordinates. Then the direct transformation from (0, A) to (X, Y) is given
by
X = koN^A + (1 - T+QA3 1 6
+ (5 - 1 8 T + T2 + 72C - 58e'2)A5 1 1 20], (4.233-5)
Y = koiSt - %, + JV0 tan 0[A2 / 2 + (5 - T + 9C + 4C2)A4 I 24
+ (61 - 58 7>r2+600C-330£?'2)A6/720]}. (4.233-6)
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
The scale factor at any given point is given by
213
k = h
1 + (1 + Qy + (5 - 4 T + 42C + 1 3C2 - 28e'2)^
+ (61 - 148 7+ 16 T2)
720
(4.233-7)
The inverse formulas, for converting (X, Y) to (0, A) are
p 0! - {Nfr tan 0, / ff ,)[£>2 12 - (5 + 3 Tt + 10d - 4C2 - 9ea)D4 1 24
+ (61 + 90 Ti + 298C, + 45 T2 - 252e'2 - 3C2)D6 / 720], (4.233-8)
A = Aq + [£>-(! + 2T\ + Cl)D3/6
+ (5 - 2C, + 28 Ti - 3C] + 8e'2 + 24 T2)D5 1 120] / cos 0, ,
(4.233-9)
where 0j is the foot-point latitude or the latitude at the central meridian which has
the same Y coordinate as that of the point (0, A). It is given by
0, ^M + (3e1/2-27e:J/32 + -)sin2Ai + (21e2/ 16-55e4/32- --)sm4A*
+ (151ef /96+ -)sin6/x + ■
(4.233-10)
where
i - vT^2
1 +VTre2
# = S![a{\ - e2 1 4 - 3e4 / 64 - 5e6 I 256 - •■■)],
S = S4>n + Y/k0.
(4.233-11)
(4.233-12)
(4.233-13)
For the other terms, in addition to needing C, T, and N$ for the footpoint latitude
01 we also have
Ri = a(l -e2)/(l - e2 sin2 0O3'2,
D = X/(N^kn).
For UPS coordinates, the direct equations are
(4.233-14)
(4.233-15)
X - 2000000 + p sin A,
ynorth = 2000000 - pcosA,
ySOmh = 2000000 + pcos A.
(4.233-16)
(4.233-17)
(4.233-18)
214 EXPLANATORY SUPPLEMENT
The first Y is for the north zone, the second for the south zone. The auxiliary
quantities are obtained from
^(1 +e),+e(l - e)'-e*
f = tan
We also have for the scale
where
7T (j)
4~2
k =
1 — e sin <f>
1 + e sin 4>
en
am
cos 4>
in — . ■ ■ .
sj 1 — e1 sin2 4>
The inverse solution for cf> and A is given by
<b= arctan I t
2
I — e sin i
L + e sin 4>
A - arctan
-Y
(4.233-19)
(4.233-20)
(4.233-21)
(4.233-22)
(4.233-23)
(4.233-24)
4.234 State Plane Coordinates In the United States, a state plane coordinate sys-
tem has been established so that each state has its own grid coordinate system.
These systems were originally designed by the U.S. Geological Survey in the 1930s
and 1940s, and have since been legally adopted by the state legislatures of most
states. For each state, they consist of one or more zones of transverse Mercator for
states or areas of states with a mostly north-south extent or Lambert conformal
conic projections for states or areas with mostly east-west extent (and one oblique
transverse Mercator projection for part of Alaska). There are 136 separate zones
covering all of the 50 states (the U.S. Virgin Islands, American Samoa, etc., all
also have their own projections, some of which are not conformal). Snyder (1983,
pp. 58-62, Table 8) lists all the zones, by projection and state, with the central
meridian, its scale error for the Lambert conformal conic zones, and the standard
parallels and origin for the transverse Mercator zones. The coordinates of the ori-
gins are also given in a footnote to that table. Note that survey feet (where 1 m =
39.37 in.) are used rather than standard U.S. feet (where 2.54cm = 1 in.). These
systems were established under the assumption that any calculations done assum-
ing plane coordinates would not have errors worse than 1 in 10000. Their definition
in many states is also legally tied to the North American Datum of 1927 (see the
next section), and land parcels may be defined or even must be defined in terms of
state plane coordinates. It seems likely that the use of state plane coordinates may
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
215
fall off in the future since: (a) a 1 in 10000 error in surveying urbanized areas is
noW becoming unacceptable, and (b) the NAD1927 has been replaced for most pur-
noses by the North American Datum of 1983, and the state plane systems will now
need to be updated-a difficult procedure since it involves changing laws in several
states However, many governmental and commercial organizations currently use
state plane coordinates heavily. [For more information on state plane coordinates
and projections, see Mitchell and Simmons (1945) and Snyder (1983, pp. 56-63).
General surveying texts, such as Moffit and Bouchard (1975), also contain informa-
tion on state plane coordinates.]
The equations for the transverse Mercator projection are already given under
the description for UTM above. The only changes will be for constants such as the
central meridian (A0), the scale factor on the central meridian (k0), and the gnd
coordinates of the origin.
The equations for the Lambert conformal conic projection are taken from Sny-
der (1983, p. 107-109). The direct equations are
X = psinfl,
Y- p0 - pcos0,
k ,
am
(4.234-1)
(4.234-2)
(4.234-3)
where
P =
Pa =
aFt",
n(\ — Ao),
oF<g,
In m\
lnm2
ln?i - Inf2
COS (j)
n= (1 -^sin2^)"2'
? = tanU-2j
mi
sin
sin 0)
en •
<
(4.234-4)
(4.234-5)
(4.234-6)
(4.234-7)
(4.234-8)
(4.234-9)
(4.234-10)
with the same subscripts 1, 2, or none applied to m and 4> in the equation for
the subscripts 0, 1, 2, or none applied to / and * in the equation for t. n, F, and po
need be computed only once per zone. Note that published (in some cases officaily
adopted) tables may give (negligible) differences, as the tables were computed Witl
simpler, less-precise formulas (Snyder, 1983, p. 108).
216
The inverse formulas are
EXPLANATORY SUPPLEMENT
7T
2
2 arctan
1 — esin <j>)
1 + esin 0)
tti
where
' = (^)"n,
aF
p = ± s/x1 - (po - m
A = - + A0,
n
9 = arctan ■
Po-Y
(4.234-11)
(4.234-12)
(4.234-13)
(4.234-14)
(4.234-15)
The sign of p is taken from n. If n is negative, the signs of X, Y, and po in the
equation for 6 must be reversed, and the quadrant of Q determined accordingly.
Also note that the equation for cf> requires iteration on 0, and will converge rapidly
with the initial assumption
<t> =
2 arctan t.
(4.234-16)
4.235 Other Grid Systems Obviously, the equations that have been given above
for the Transverse Mercator, Polar Stereographic, and Lambert Conformal Conic
projections can be used with many other grid systems with just a change of the
appropriate constants.
4.24 Geodetic Datums and Reference Systems
A geodetic datum or reference system provides a framework to which the coordi-
nates of any point may be referenced. A geodetic datum has a very specific definition
in terms of classical geodesy; however, some recent "datums" and most terrestrial
reference systems do not fall strictly under this definition.
A geodetic datum may also be either a horizontal or vertical datum, differenti-
ating whether the datum is to serve primarily for determining (horizontal) location
on or near the surface of the Earth, or for determining height (actually gravity
potential) of a given point. In the discussion that follows, it is a horizontal datum
that is being referred to.
In this section, we briefly describe how these ellipsoids, datums, geodetic ref-
erence systems, and terrestrial reference systems are created, and then list many
of the ones in use. Information on how transformations may be performed between
these various datums and reference systems is also provided. Finally, a short expla-
nation is given of how to reduce common geodetic and astronomical observations
to the reference ellipsoid.
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 217
4.241 Datums, Geodetic Reference Systems, and Terrestrial Reference Systems A
geodetic datum is a coordinate system defined relative to a reference ellipsoid (as
defined in Section 4.21) and an initial point (or origin) near the surface of the ellip-
soid. Therefore the following are needed: (a) the size and shape of the (rotational)
ellipsoid (e.g., a, 1 //); (b) the astronomical coordinates of the initial point of the
ellipsoid ($,A), and an azimuth (a) determined from that point; (c) deflections
of the vertical and a geoid undulation at the initial point. Obviously an infinite
number of choices are possible for the above parameters, but several methods are
usually applied in order to provide the most useful possible reference system. As
previously described, the ellipsoid is chosen so as to represent the geoid as well as
possible over the desired area (e.g., a country) or the entire Earth. For a local da-
tum, the deflections and undulation are sometimes chosen for simplicity to be zero
at the initial point — but more usefully over larger areas or for a terrestrial datum,
these values are chosen so that the rotational axis of the ellipsoid will be parallel
to or identical with that of some reference rotational axis of the Earth (and if the
deflection components are absolute, then the datum is also geocentric).
As an historical note, it should be added that classical reference ellipsoids and
geodetic datums were usually established in support of the surveying of great hor-
izontal geodetic networks, using triangulation networks over large areas as well as
auxiliary baseline, astronomical, and gravity measurements. The data from these
surveys could be adjusted mathematically in order to specify the best-fitting ellip-
soid to the area being surveyed, as well as the other desired parameters for a datum.
An initial point for the datum was usually chosen near the center of the network
in order to minimize the errors in measurements relative to the initial point as one
moved outward in the network.
However, the space age and its introduction of artificial satellites greatly changed
the methods of datum determination. Now the absolute positions relative to the
center of the Earth could be determined for various points in the geodetic network
(the techniques used for this are described in Section 4.4). These position determi-
nations could be combined with the conventional surveying observations in order
to obtain much more precise datums for much larger areas. Further, coordinate
systems (usually with sets of station coordinates) can be determined without refer-
ence to any ground-based surveying observations or even to an ellipsoid (although
an ellipsoid is often used to express station coordinates in the more conventional
geodetic latitude and longitude instead of geocentric cartesian coordinates). The
orientation and scale of these space-based systems is usually obtained from some
combination of holding the coordinates of one or more stations fixed, and using or
determining an Earth rotation series that fixes the changes in the system over time
due to the Earth's rotation. In practice these assumptions are made so that the
new system will have the same orientation and scale as some previous "standard"
system, such as the BIH or IERS systems (described below). The center of the
Earth is determined dynamically by the satellite observations.
218 EXPLANATORY SUPPLEMENT
One of the first consequences of the availability of the new techniques was
the establishment of geodetic reference systems that have much in common with
a geodetic datum but have slight differences in their definition. Such a reference
system does not rely on the definition of an initial point. The orientation of such
a system is assumed to be "perfect," i.e., with rotational axis, reference meridian,
and center, being that of the Earth. The reference ellipsoid in this case is always
a terrestrial reference ellipsoid, and the best possible representation of the mean
Earth ellipsoid, or that ellipsoid that best approximates the geoid. To accomplish
the latter, the gravity potential at the surface of the ellipsoid is specified (or al-
ternatively, J2, the dynamical form factor of the Earth) as well as the rotational
velocity of the Earth. The constants and parameters of such a system are of course
always undergoing refinement, but the specification of such systems allows for their
use in referencing geodetic observations.
A final refinement beyond a geodetic reference system is a terrestrial reference
system., using a reference frame consisting of a set of cartesian station coordinates
and their velocities to provide the underlying definition rather than a reference
ellipsoid. Many other parameters and even methods of data reduction ( "Standards" }
are also given as part of the reference system — generally in order to define how
observations will be used in order to define the dynamic connection between such a
terrestrial system and a celestial one. By international agreement, one such system
is defined as fundamental and it is the IERS Terrestrial System, described further
in Section 4.2.5.
4.242 Specific Ellipsoids, Datums, and Reference Systems Table 4.242.1 lists var-
ious ellipsoids that have been used as the basis of geodetic datums and reference
systems. Note that as previously mentioned, only a and 1 / / are necessary for the
definition of an ellipsoid. Table 4.242.2 lists geodetic datums (also giving the origin
station if any and its latitude and longitude) and notes that are associated with spe-
cific terrestrial reference systems. Similar, but abbreviated versions of these tables
are given in the current Astronomical Almanac, page K13. Finally, Figure 4.242.1
(after Rapp, 1980, p. 82) shows on a world map t he coverage of the major geodetic
datums.
4.243 Datum Conversions The subject of converting coordinates from one datum
to another is a complex one, with a very extensive literature. Indeed, the problem
almost always arises and is usually discussed whenever a new datum or reference
system is formed, since there will usually be a need to transform the coordinates of
points in other systems to coordinates in the new system.
Two general methods have been used in the past for datum transformations.
The first is a differential method, where differential (in the partial-derivative sense)
equations are formed that give the change in station coordinates (A0, AA) as a
function of changes in datum/ellipsoid parameters (such as for changes in a, f,
Figure 4.242.1
Major geodetic datum locations
220
Table 4.242.1
Earth Ellipsoids
EXPLANATORY SUPPLEMENT
bllipsoiu
Semi-major
Inverse
(name and year computed)
axis (a) in
flattening
II It? It? 15
V I ]}
Airy 1830
6377563.396
299.324964
Everest 1830
6377276.345
300 8017
Everest 1830 (Boni alternate)
6377301.243
300 8017
Bessel 1841
6377397.155
299.152813
Clarke 1866
6378206,4
294.978698
Clarke 1880
6378249.145
293 465
Clarke 1880 (modified)
6378249.145
293.4663
Helmert 1906
6378200
298.3
International 1909
6378388
297
Krassovsky
6378245
298.3
Mercury 1960
6378166
298.3
WGS60*
6378165
298.3
IAU 1964
63781 60
298.25
Australian National 1 965
6378160
298.25
WGS66*
6378145
298.25
Modified Mercury 1968
6378150
298.3
South American 1969
6378160
298.25
Geodetic Reference System 1 967"
63781 60
298 2471 674P7?
WGS72*
6378135
298.26
lAf^ 1 Q7C
6378140
298.256
IAU 1976
6378140
298.257
Geodetic Reference System 1 980'
6378137
298.257222101
MERIT 1983*
6378137
298.257
WGS84*
6378137
298.257223563
IERS 1989*
6378136
298.257
* Ellipsoid is defined or recommended as part of a geodetic or terrestrial reference
system,
Sources: Rapp (1980, p. 83), Astronomical Almanac (1 990, p. K130), DMA (1987,
part 1, p. 3-41), McCarthy (1989).
scale, etc.). This type of method has been most useful in the past where small
differences exist between the two datums, or a minor change in an existing datum
was being made. However, the derivation of the differential equations for a specific-
case and their application and accuracy is a complicated subject and will not be
discussed further here. [Information on this type of method is available in Rapp
(1980, pp. 42-74); Bomford (1971, pp. 199-207); Heiskanen and Moritz (1967, pp.
204-210); Molodenskii et al. (1962, pp. 13-17); Torge (1980, pp. 150-152); DMA
(1987, part 1, pp. 7-1 to 8-8).]
Table 4.242.2
Geodetic Datums
Datum
Ellipsoid
Origin
I atitiiHo
[ nnnttiiHp
LUI IUILUUC
Adindan
r"\ar\ia 1 Rdf\
\j\a'n.G ItJtJU
^tatinn 7*.,
OLdLILJI I £- \f
+22 1 0 07. 1 1 0
0312921.608
American odniud isot
Clarke 1866
Betty 1 3 ECC
-142008.341
1891707.750
Arc- Cape \ooum wncd;
Clarke 1880
Buffelsfontein
-335932.000
0253044.622
Argentine
International
Campo Inchauspe
-355817
297 49 48
Ascension Island 1958
International
Mean of 3 stations
-0757
34537
Australian Geodetic 1966
Australian National
Johnston Memorial Calm
-255654.55
1331230.08
Bermuda 1957
Clarke 1866
Ft. George B 1937
+322244.360
2951901.890
Beme 1 898
Bess&l
Berne Observatory
+465708 660
0072622 335
Betio Island, 1966
International
1 966 Secor Astro
+01 21 42. 03
1725547.90
Camp Area Asiro i sro i f uo<jo
International
Camp Area Astro
-775052.521
1664013.753
p.ntnn Antrn 19R6
International
t966 Canton Secor Astro
-024628.99
1B8 1643.47
Cape Canaveral {")
Clarke 1866
Central
+282932.364
2792521.230
Christmas Island Astro 1967
Intel national
Sat.Tri.Sta. 059 RM3
+020035.91
2023521.82
Chua Astro (Brazil-Geodetic)
Internationa!
Chua
-194541.16
3115352.44
Corrego Alegra {Brazil- Mapping)
International
Corrego Alegre
-195015. 140
3110217.250
raCtAr iclsnd 1 Qfi7 Astro
International
Satrig Rm No. 1
-271039.95
250341681
Efate (New Hebrides)
International
Belle Vue IGN
-174417.400
1682033,250
European (Europe SO)
International 1924
Helmertlurm
+522251.45
0130358.74
Qraciosa Island (Azores)
I nternational
Sw Base
+390354.934
3315736.118
Gizo, Provisional DOS
I nlrrn^linnal
I HIV lallLFI lal
GUX 1
-092705.272
1595831.752
Guam
Clarke 1 866
Togcha Lee No. 7
+132238.49
144 4 5 51.56
Heard Astro 1969
International
Intsatrig 0044 Astro
-5301 11.68
0732322.64
Iben Astro, Navy 1947 (Truk)
flarlro 1 ftfifi
Iben Astro
+072913.05
151 4944.42
Indian 1938
FuAra^t 1RTC1
Kalianpur
+240711,26
0773917.57
Isla Socorro Astro
riarko "Iftfifi
L-lpi ^t;: IOvU
Station 038
+1843 4493
2490239.28
Inhnctnn Iclanrl
JUIlllblUII ICjIqI IU
International
Johnston Island 1961
+1644 49.729
1902904.781
Kourou (French Guiana)
International
Point Fondamental
-051553 699
-524809. 149
Kusaie, Astro lytu, lyco
International
Alien Sodano Light
+0521 48.80
1625803.28
Luzon 191 1 (Philippines)
^lanvy i odd
Ba I a n ca n
+133341.000
1215203.000
Midway Astro 1961
1 nternational
Midway Astro 1961
+28 1 1 34. 50
1623624.28
New Zealand 1949
International
Papalahi
-41 1908.900
1750251.000
North American 1927
\_>ldl Kfc I ODD
Meades Ranch
+39 1326. 686
26127 29.494
North American 1983
rspi Rn
una du
Old Bavarian
Bessel
Munich
+480820. 000
0113426.483
Old Hawaiian
/-> — r|,Q -H QfiC
i^jarKe iodd
Oahu West Base
+21 1813. B9
2020904.21
OrHiisfir-P* ^uruau f~i R 1Q1R
\J 1 \Jl ILL- OUIVLV VJ.L> \ O 0*J
Airy 1830
Herstmonceux
+5051 55.271
0002045.882
OSGB 1970 (SN)
Airy
Herstmonceux
+5051 55. 271
0002045.882
Palmer Astro 1969 (Antarctica)
International
ISTS 050
-644635.71
2955639.53
Pico de las Nieves (Canaries)
international 1 3tn
Pico de las Nieves
+275741.273
3442549.476
Pitcairn Island Astro
Into mat ions I
Dili-aim Aclrn 1 QfiT
—250406. 97
2295312.17
Potsdam
Helmertlurm
+522253.954
0130401 153
Provisional S, Amer can 1966
International
La Canoa
+0834 17. 17
2960825. 12
Provisional S. Chile 1963
International
l-4ifn Will
n no a v 1 1 1
—535707. 76
2912328.76
Pulkovo 1 942
KrassovsKi iy4^
+5946 18. 55
030 1942.09
Qomoq (Green la nti)
International
No. 7008
South American 1969
South American 1969
Chua
-194541 653
31 1 5355. 936
ftm ithd-Sft Irh-H /kiik-.i-t\
ouumeasi isiano (ivianej
r^ifsrko mnn
L<iarKe i ddu
-04 4039.460
0553200. 166
Smith cm f n t !i Aelr^i
Inte mational
ISTS 061 Astro Point 1968
-54 1638.93
3233043.97
Swallow Islands (Solomons)
International
-101821 42
1661756.79
Tananarive
International
Tananarive Observatory
-185502. 10
04733 06 75
Bessel 1941
Tokyo Observatory (old)
+3539 17.51
13944 40. 50
Tristan Astro 1968
International
Intsatrig 069 RM No. 2
-370326.79
3474053.21
VitiLevu 1916 (Fiji)
Clarke 1880
Mfinavatu (latitude only)
— 1753<?o. £od
Wake Island, Astronomic 1952
Suva {longitude only}
1782535.835
International
Astro 1 952
+191719.991
1663846.294
White Sands (*)
Clarke 1866
Kent 1909
+323027.079
2533101.306
Yof Astro 1967 (Dakar)
Clarke 1830
Yot Astro 1967
+144441. 62
3423052.98
* Local datums of special purpose, based on NAD 1 927 values for 1he origin stations.
Sources: Rapp (1980), Mueller eta/. (1973. p. 11), Astronomical Almanac (1990, p. K13).
222 EXPLANATORY SUPPLEMENT
A second method is to do a rigorous transformation between the systems of
each datum. This is often required if high accuracy is desired, or there are several
or large differences between the datums. For any pair of datums, transformation
parameters may have already been determined, at least for offsets and sometimes
for rotation and scale. A three-step procedure is required to convert (0, A, h) from
system 1 to system 2
W,A,A)i =>(X,y,Z)i,
(X, Y,Z)i=>(X,Y,Z)2,
(X,Y,Z)2 =*(<{>, \,h)2.
(4.243-1)
(4.243-2)
(4.243-3)
The first two transformations are already described in Section 4.22. The equation
for the third transformation is
X
Y
Z
X '
" r, "
" D
-R3
R2 -
' X '
Y
D
Y
Z
l
. T3 j
. -R2
R}
D
. z .
(4.243^)
where T\ , T2, and T} are the components of a translation vector; R\ , R2, and #3 are
the rotation angles (using small-angle formulas here); and D is the excess 1 scale
factor (BIH, 1988, p. A-4). Various forms of this seven-parameter transformation
exist under various names (some not as rigorous as others). (See Kumar, 1972;
Rapp, 1980, p. 48; DMA, 1987, part 1, pp. 7-6, 7-12 to 7-14.) It is also possible that
some of the parameters may not be known or can be assumed to be zero. In modern
reference systems, D is usually close to zero, as is T\ and T2, and the rotations are
quite small.
In some cases, the transformation parameters are not well-determined or not
known at all. The user must then determine the parameters directly from observa-
tions that tie stations together between the two systems. If enough points have their
coordinates determined in both systems, Equation 4.243-4 or similar equations can
be solved (usually in a least-squares solution) for the transformation parameters.
Of course, this may not take into account any deformations of either of the datums
from their idealized reference systems (due to surveying errors). Any deformations
will be absorbed into the transformation parameters when they are solved for.
Of course, nowadays an alternative exists to doing transformations at all. One
could just measure the coordinates of all the desired points with a satellite receiver—
thus determining the coordinates of the points in some current reference system,
bypassing any transformations entirely. This eliminates the need to use the old
datums or old observations. The coordinates can be measured in an absolute sense,
relative to the center of the Earth at the 1- to 5-meter accuracy level, or relative to
other points nearby (within tens of kilometers say) to the few-cm level. The satellites
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 223
usually observed for this purpose are those of the Navy Navigation Satellite System
(NNSS) or Global Positioning System (GPS). More information on these systems
is given in Section 4.4.
4 244 Reduction of Observations to the Ellipsoid In many cases where astronomi-
cal or geodetic observations are made, it is desirable to "reduce" the observations so
that they would correspond to such observations made on the ellipsoid itself, using
the ellipsoidal normal as the local vertical. This is usually for when one is referring
observations to or densifying a current network, or for that matter creating a new
network itself.
Various formulas of various accuracies exist for this type of reduction, depending
on the type of observation involved (chord or arc distances, azimuths, angles, etc.)
and the accuracy desired. Any good surveying text will contain various versions of
these formulas. (Also see Bomford, 1971, pp. 45, 89-91, 120-122; Heiskanen and
Moritz, 1967, pp. 184-197.)
4.25 The Conventional Terrestrial Reference System
The Conventional Terrestrial Reference System is currently defined as part of the
IERS Reference System. The IERS Reference System itself consists of the IERS
Reference Frames and the IERS Standards.
The IERS Reference Frames include both a terrestrial and a celestial one, with
the terrestrial frame (ITRF) being defined with the following characteristics: The
origin, the reference directions, and the scale of the ITRF are implicitly defined by
the cartesian coordinates and velocities adopted for various "primary" observing
stations of the IERS. If geodetic coordinates are needed, the GRS 80 ellipsoid is
recommended for use in the conversion (a = 6378137 m, 1 // = 298.257222101).
The origin of the ITRF is located at the center of mass of the Earth with an
uncertainty of less than ± 10cm (as determined by laser-ranging observations).
The standard unit of length is the SI meter. The IERS Reference Pole (IRP) and
Reference Meridian are consistent with the corresponding directions in the BIH
Terrestrial System (BTS) within ± 0.005". The BIH reference pole was adjusted
to the Conventional International Origin (CIO) in 1967. It was then kept stable
independently until 1987. Recent re-reduction of the available data has shown (BIH,
1988, p. A-3) that the tie of the BTS (and therefore the IRP) with the CIO is
accurate to ± 0.03". The ITRF should show no global net rotation or translation
with time (due to the motions of the stations or the tectonic plates they lie on). A
more complete description of the IERS Terrestrial Frame is given by Boucher and
Altamimi (1989).
The Celestial Frame (ICRF) is defined similarly, using the coordinates of 23
"primary" radio sources (Arias et al, 1988). The origin is at the barycenter of the
solar system. The direction of the polar axis is given for epoch J2000.0 by the IAU
224 EXPLANATORY SUPPLEMENT
1976 Precession and the IAU 1980 Theory of Nutation (see Chapter 3). Comparison
of optical and radio observations shows that the origin of right ascension is in
agreement with that of the FK5 star catalog to within ± 0.04". The parameters
that describe the rotation of the ITRF relative to the ICRF (in conjunction with
the precession and nutation model) are the Earth rotation parameters (ERP). For
further information on celestial reference frames, see Chapter 3.
The IERS Standards (McCarthy, 1990) describes the models to be used when
data are being reduced for use by the IERS (and asks that the effects of deviations
from the standards be determined). The next section contains material substantially
drawn from those standards.
The IERS Reference System evolved from a series of previous systems. As noted
above, the ITRF is itself a direct continuation of the BIH Terrestrial System (BTS),
first established in 1985 and continued through 1987. The BTS had in turn evolved
from the BIH 1968 System, which was established in 1969 by the BIH on the basis
of a comparison of data collected by the BIH and the International Polar Motion
Service (IPMS) during the mid-1960s. However, it should be emphasized that this
was not a complete terrestrial reference system; it was a reference system for di-
rection only, being based on the observed astronomical coordinates of the involved
observing stations. The comparison of the BIH collected data with the IPMS data
allowed the BIH 1968 System to be referenced to the IPMS reference pole. This
pole is the Conventional International Origin (CIO) which was the approximate
average position of the true celestial pole of the Earth from 1900.0 to 1905.0, but
was actually defined by the adopted latitudes of 5 International Latitude Service
(ILS) stations. (For further information on all of the above, see IERS, 1989, or the
BIH Annual Reports for the appropriate year(s), or see Mueller, 1969 for further
information on these systems for the pre-1969 period.)
4.3 GRAVITY, THE TIDES, AND MOTIONS OF THE CRUST
In this section, various phenomena will be covered that affect the Earth's gravity
field and coordinates of a given station. A very short summary of one method of
modeling the Earth's gravity field (as a spherical harmonic series) is given, along
with the IERS recommendations for a "standard" gravity field. This is followed by,
almost verbatim, the IERS Standards recommendations on Earth tides, ocean tides,
ocean and atmospheric loading, and plate motions (MCarthy, 1989). Finally, the
effect of tides on the Earth's rotation itself is described, as given by the Standards.
4.31 Modeling the Earth's Gravity Field
The Earth's gravity can be considered as a force that is the sum of the Earth's
gravitational force and the centrifugal force resulting from the Earth's rotation.
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 225
The Earth's external gravitational field can be considered a potential field
whose strength V fulfills the Laplace equation
AV = 0, (4-31-1)
where A is the Laplacian operator
+ + (4.31-2)
dX2 dY2 8Z?
in a cartesian X, Y, and Z coordinate system.
It is possible to develop a solution to Equation 4.31-1 (Heiskanen and Moritz,
1967. pp. 15-35; Torge, 1980, pp. 26-29) in terms of a series of Legendre polynomials:
PM)=—i—x^(t2-ir (4-31-3)
with .. j.
t = cosi/> = cos0cos0' + sin0sinl9'cos(A - A). (4.31-4)
xji is the spherical distance on a unit sphere between an attracted point (at spherical
coordinates r, 9, A) and an attracting point (at r', 8', A').
In order to express Pn(t) as a function of the spherical coordinates 8 and A, it
may be decomposed using associated Legendre Functions of the first kind of degree
n and order m, obtained by differentiating P„U) m times
E." p-M-v-er^PM- (431-5)
It can then be expressed by the functions
Pnm(cos0)cosmA, (4.31-6)
P„m(cos0)sinmA, (4.31-7)
known as surface spherical harmonic functions.
A general equation for the gravitational potential then becomes
- _ X^X" k ~ ;
- r Z^t 2—i („ + m)l
n=0 m=0
x I ( pnffl(c0s 6) cos mX / / f r"'P,„r,(coS 8') cos mXdm
f\ J J ./Earth
+ P^Ccqs 8) sin mX fff r'"P,„H(cos 8') sin mX dm
J J ./Earth
(4.31-8)
226 EXPLANATORY SUPPLEMENT
1 for m - 0
2 form t'O '
jfc =
If the spherical component of the potential of the Earth's mass M is separated
out (with n = 0), the ellipsoidal semi-major axis a is used to provide a scale for the
unit sphere, and the integrals are replaced by harmonic coefficients, Cnm and Snm,
we then have
GM
T
1 + X! ]L (j) (C™ C0S mA + Snm Sin ^PnmiCQS 9)
n= I m=0
(4.31-9)
In many cases, an alternative notation is
Jn = -C„, (4.31-10)
]m = -Cmn, (4.31-11)
Knm= S„m. (4.31-12)
It may be convenient to use fully normalized surface spherical harmonics:
Pnm(cos 9) = Jk(2n + l)(-^LJ^pnm(cos e), k= J J fTm-° (4.31-13)
y (« + m) ! [2 form^O
This results in fully normalized coefficients
f C„n 1 / (n+fH)! ~
I Snm J Y*(2n + O(n-m)!
It should be noted that the expansions for V converge rigorously only outside a
sphere of radius a that just encloses the Earth, so that the Laplace equation may be
satisfied. In practice, however, since the coefficients are normally determined em-
pirically from measurement (e.g., from satellite observations), the series can always
be considered to converge. Once a set of coefficients are determined (see the next
subsection) the expression for V becomes valid and useful for anywhere in space.
The same expression may be used for determining the gravitational potential
on the Earth's surface. However, as noted previously, it is the gravity potential, here
denoted W (including centrifugal force) that is of primary interest on or near (i.e.,
moving with) the Earth's surface. We have
jfc =
for m = 0
for m 7/ 0
(4.31-14)
W= V+l-i4{X2 + Y\
(4.31-15)
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 227
where wE is the angular velocity of the Earth, and X and Y are the coordinates of
the point in question (P(X, Y,Z)).
It should be noted that many other representations of the Earth's gravity field
exist and are widely used, depending on the application. Most commonly, W may
be determined over a wide area or the Earth itself, using measured gravity-related
data. Any of the standard references previously mentioned on geodesy and geodetic
astronomy contain detailed information on these types of computations.
4.32 A Representation of the Earth's Gravity Field
The current IERS recommended geopotential field is the GEM-T1 model given in
Table 4.32.1. The GM and a values reported with GEM-T1 (398600.436 km3 /s2 and
6378137 m) should be used as scale parameters with the geopotential coefficients.
The number of terms actually needed to provide sufficient accuracy for a given
application will, of course, vary. For example, when used in computing the orbit of
the LAGEOS satellite, only terms through degree and order 20 are required.
Values for the C2i and S2i coefficients are not included in the GEM-T1 model
(they were constrained to be zero in the solution), and so they should be handled
separately.
The C21 and S21 coefficients describe the position of the Earth's figure axis
with respect to the ITRF pole. When averaged over many years, the figure axis
should closely coincide with the observed position of the rotation axis (or "mean
pole") averaged over the same time period. Any differences between the mean figure
and mean rotation axes averaged would be due to long-period fluid motions in the
atmosphere, oceans, or Earth's fluid core (Wahr, 1987, 1990). At present, there is
no independent evidence that such motions are important; so it is recommended
that the mean values used for C21 and S2i give a mean figure axis that corresponds
to the mean pole position.
The BIH Circular D mean pole positions from 1982 through 1988 areconsistent
to within + 0. 0005" corresponding to an uncertainty of ± 0. 01 x 10~9 in C21 (IERS)
and 52, (IERS).
If the mean pole during this period coincided with the ITRF pole, then C2i =
S21 ~ 0 could indeed be used. However, the poles are offset by the angular displace-
ments x and y, so that
C21 = XC2Q,
S21 = — yC2o
(4.32-1)
(4.32-2)
Table 4.32.1
GEM-T1 Normalized Coefficients (x 106)
Zonals
Index
Value
Index
Value
Index
Value
Index
Value
ndex
Value
n
m
n
m
n
m
n
m
n
m
•2
0
-484.1649906
3
0
0.9572357
4
0
0.5387322
5
0
0,0687802
6
0
-0.1481004
7
0
0.0905337
8
0
0.0459023
9
0
0.0283764
10
0
0.057221 1
11
0
-0.0512619
12
0
0.0320806
13
0
0.0422319
14
0
-0.0197327
15
0
0.0018731
16
0
-0.009377a
17
0
0,0203968
18
0
0.0112912
19
0
-0.0046084
20
0
0.0153150
21
0
O.OO97754
22
0
-0.004844D
23
0
-0.0241260
24
0
-0.0009556
25
0
0.0068867
26
0
00018391
27
0
0,0041234
28
0
-0.0058541
29
0
-0.0039091
30
0
-0.0002749
31
0
0.0051154
32
0
0.0000819
33
0
0.0022286
34
0
-0.0024803
35
0
0.0012731
36
0
0.0007396
Sectorials and Tesserals
Index Value Index Value Index Value
n m C S n m C S n m C S
" 3
2 0297737
0.2495946
6
1
-0.0813751
0.0238900
9
1
0.1480477
0.0245251
12
1
-0.0492610
-0.0496520
15
1
0.0082868
0.0142124
18
1
-0.0002253
-0.0456055
21
1
-0.0153942
0.0417459
24
1
0.0081178
-0.0291987
27
1
0.0005230
0.0066113
30
1
-0.0016171
-0.0090886
33
1
0.0012672
0.0021258
36
1
0.0028774
-0.0058408
4
2
0.3470021
0.6640304
7
2
0.31 771 08
0.0916083
10
2
-0.0805212
-0,0513356
13
2
0.0534361
-0.0575844
16
2
-0.0156437
0.0245431
19
2
0.0084369
-0.0104744
22
2
-0.0142925
0.0020958
25
2
0.0037220
0.0052068
28
2
-0.0084276
-0.0115524
31
2
0.0066008
0.0008362
34
2
0.0035817
0.0051783
3
3
0.7209866
1.4131694
6
3
0.0619709
0.0046430
9
3
-0.1553742
-0.0840158
12
3
0.0324198
0.0179438
15
3
0.0446271
0.0265447
18
3
-0.0010020
-0.0070483
21
3
0.0019941
0.0226923
,1
1
-0 5334272
-0,4751 189
7
1
0.2770971
0.0978177
10
1
0.0769655
-0.1381110
13
1
-0.0540617
0.0434555
16
1
0.0317099
0.0173493
19
|
-0.0115942
0.0053764
22
1
0.0083946
-0.0147250
25
1
0.0037145
0.0043498
26
1
0.0065294
-0.0100251
31
1
0.0051782
0.0023829
34
1
-0.0015608
-0.0091011
2
2
2,4389280
-1.3998397
5
2
0.6557902
-0.3234056
8
2
0.0703801
0.0684494
11
2
0.0090541
-0.0992414
14
2
-0.0348122
-0.0060681
17
2
-0.0057800
0.0171247
20
2
0.0198772
0.0032259
23
2
-0.0005313
-0.0017780
26
2
-0.0052887
0.0002522
29
2
0.0094632
-0.004331 1
32
2
0.0018649
0.0043749
35
2
-0.0025529
0.0010277
4
3
0.9909779
-0.2006215
7
3
0,2507429
-0.2091639
10
3
-0.0013119
-0.1614824
13
3
-0.0140259
0.0836615
16
3
-0.0320841
-0.0450272
19
3
0.0014391
0.0141955
22
3
0.0067253
-O.008O913
5
1
-0.0589503
-0.0955435
8
1
0.0288561
0.0547223
11
1
0.0095019
-0.0278111
14
1
-0.0187462
0.0232244
17
1
-0.0309381
-0.0268459
20
1
0.0145119
-0.0212711
23
1
0.0008657
0.0145970
26
1
0.0049741
-0.0172518
29
1
0.0034699
0.0024176
32
1
-0.0091529
-0,0092451
35
1
-0.0019069
0.0020110
3
2
0.9035491
-0.6204198
6
2
0.0516096
-0.3749956
9
2
0.0311365
-0.0323882
'2
2
0. 0076400
0,0349183
15
2
-0.0216258
-0.0364425
18
2
0.0084083
0.0168428
21
2
0.0009874
-0.0026067
24
2
-0.0058515
0.0052022
27
2
0.0102174
-0.0028223
30
2
-0.0040515
-0.0053645
33
2
-0.0010231
00009548
36
2
0.0001 790
0.0012922
5
3
-0.4482036
-0.2151363
8
3
-0.0199664
-0.0869367
11
3
-0.0288895
-0.1324963
14
3
0.036931 1
0.0224222
17
3
0.0101214
0.0098939
20
3
0.0082691
0.0137151
23
3
-0.0045642
-0.0119458
Table 4.32.1, continued
GEM-T1 Normalized Coefficients (x 106)
Index Value Index Value Index Value
n m~ C S n m C S n m C s
24
3
0.0069148
-0.0105925
25
3
-0.0032642
-0.0031087
26
3
-0.0002621
-0.0037215
27
3
-0.0051035
-0.0018585
28
3
-0.0002646
0.0011662
29
3
-0.0043864
-O.O0 17609
30
3
—0,0016075
0.001 4770
31
3
— 0.0018668
—0.0040769
32
3
— 0.0006663
0.0029992
33
3
-0.0017072
—0.0028776
34
3
— 0.0005145
0.0022065
35
3
0.0006733
0.0006861
36
3
-0.0008129
—0,0013852
4
4
—0.1900348
0.3084595
5
4
— 0.2948236
0.0524087
6
4
-0.0927975
-0.4733069
7
4
-0.2737404
-0.1220207
8
4
-0.2460639
0.0677453
9
4
-0,0128303
0.0232637
10
4
-0.0973123
-0.0693825
11
4
-0.0332108
-0.0700036
12
4
-0.0653020
-0.0030125
13
4
—0.0088182
-00003732
14
4
-0,0088329
0.0018783
■5
4
-0,0443760
0.0126416
16
4
0.0365123
0.0438559
17
4
0.01 25878
0.0312242
16
4
0.0434167
0.0060924
19
4
0.0025699
0,0076747
20
4
-0,0017951
0.0008281
21
4
-0.0002550
0.0069894
22
4
-0,0094462
0.0167100
23
4
-0.0100336
-0.0016696
24
4
0.0060580
0.0181145
25
4
0.0063503
-0.0015335
26
4
0.0053222
0.0048968
27
4
0.0029164
-0.0006401
28
4
0.0028786
-0.0024855
29
4
-0.0063641
0.0007289
30
4
-O.0021 102
-0.0039250
31
4
-0.0049509
—0.0016392
32
4
0.0020179
-0.0033825
33
4
-0.0002362
0,0003938
34
4
0.0028502
-0.0018773
35
4
0.0027442
0.0014864
36
4
0.0001181
-0.0058408
5
5
0.1777563
-0.6660281
6
5
-0.2657650
-0.5377472
7
5
0.0034750
0.0196519
8
5
-0.0249335
0.0853003
9
5
-0.0141122
-0.0600627
10
5
-0.0504401
-0.0438269
11
5
0.0459086
0.0552848
12
5
0.0306040
-0.0014745
13
5
0,05961 76
0.0574460
14
5
0.0227952
-0.01 16078
15
5
0.0160742
0.0108864
16
5
-0.0077242
-0.0016773
17
5
-0,01 1 1472
-0.0056031
1B
5
0.0017426
0.021 1276
19
5
-»0.0024457
0.0173789
20
5
-0.0104182
0.0003034
21
5
0.0177593
-0.0158911
22
5
-0.0046335
-0.0001257
23
5
0.0019941
-0.0079407
24
5
-0.0140838
-0.0079805
25
5
-0.0024585
-0.0023514
26
5
0.0043007
0.0106472
27
5
-0.0015884
0.0037103
28
5
0.0029222
-0.0002361
29
5
0.0034464
0.0035422
30
5
0.0033916
0.0006983
31
5
0.0009096
0.0014649
32
5
-0.00004BO
-0.0027949
33
5
-0.0005132
0.0032353
34
5
-0.0012952
0.0000401
35
5
-0.0002351
-0.0007374
36
5
-0.0011798
0.0003403
6
6
0.0090593
-0.2363344
7
6
-0.3578527
0.1509175
8
6
-0.0664178
0.3128323
9
6
0.0705263
0.2166285
10
6
-0.0347366
-0.0777189
11
6
0.0084723
0.0242910
12
6
0.0013881
0.0458322
13
6
-0.0223869
-0.0116360
14
6
-0.0031868
0.00651 19
15
6
0,0272318
-0.0517077
16
6
0.0179949
-0.0267835
17
6
0.0002920
-0.0204166
18
6
00311991
-0.0085569
19
6
-0.0062545
0.0039169
20
6
0.0127607
0.0009662
2'
6
0.00421 46
-0.0083501
22
6
0.0146261
0.0024348
23
6
0-0099988
0.0049568
24
6
-0.0003055
-0.0006641
25
6
0.0059358
-0.0067430
26
6
0.0085388
0.0031968
27
6
0.0017746
-0.0021342
28
6
-0.0083296
0.0020029
29
6
-0.0002960
-0.0024663
30
6
-0.0032177
0,0041679
31
6
-0.0005622
0.0007667
32
6
-0.0038310
0.0002190
33
6
0.0013824
-0.0012542
34
6
0.0005808
-0.0003220
35
6
0.0008304
-0.0014752
36
6
-0.0005757
-0.0008988
7
7
0.0015976
0,0220013
8
7
0.0704248
0.0748626
9
7
-0.1186233
-0.1005510
10
7
0.0097468
-0.0042901
11
7
0.0096093
-0.0918891
12
7
-0.0126975
0.0348291
13
7
0.0035736
-0.0066171
14
7
0.0374843
- 0.0043588
15
7
0.0667130
0.0114545
16
7
0.003051 1
-0.0090737
17
7
0.0229520
-0.01 19809
IS
7
-0.0007957
0 0067159
19
7
0.0051443
-0.0016620
20
7
-0.0077913
0.0048958
21
7
-0.0122279
-00014058
22
7
0.0127538
0.0013042
23
7
-0.0023282
0.0026452
24
7
-0.0025113
0.0050662
25
7
0.0002632
0.0034596
26
7
0.0054524
0,0025831
27
7
0.0069279
-0.0028217
28
7
-0.0046459
-0.0015444
29
7
0.001 1 830
-0.0072940
30
7
-0.0001720
-0.0001291
31
7
0.0014817
-0.0017479
32
7
-00030832
0.0018611
33
7
-0.0000405
0.0017756
34
7
0.0023148
0.0001028
35
7
0.0000748
0.001 5282
36
7
-0.0001662
-0.0004199
8
8
-0.1188827
0.1223320
9
8
0.1844954
-0.0018494
10
8
0.0437468
-0.0924808
11
8
-0.0063530
0.0225827
12
8
-0.0212177
0,0169046
13
8
-0.0122964
-0.0110925
14
8
-0.0329416
-0.0131814
15
8
-0.0406660
0,0247325
1 6
S
-0.0134376
0.0022804
17
8
0.0311564
0.0087750
18
8
0.0457191
0.0004305
Table 4.32.1, continued
GEM-T1 Normalized Coefficients (x 106)
Index Value Index Value Index Value
17
Q
s
fn
c
S
19
g
0.0148626
-0.01 13273
20
6
-0,0020109
-0,0012912
21
8
-0.0181008
0.0025208
22
e
—0.0098173
-0.0068267
23
8
0.0042209
-0.0067954
24
8
-0.0024330
0.0075574
25
5
0.0014256
-0.0041241
26
8
0.0030889
-0,0021407
27
8
-0.0041794
-0.0044059
28
8
-0.0005655
-0,0031778
29
8
-0.0064245
0,0025406
30
8
0.0029536
0.0005720
31
8
0.0000872
-0.0011845
32
8
0.0008818
0.0033226
33
8
-0,0000823
0.0015750
14
Q
-0.0007860
35
8
0.0002679
0.0001772
36
8
-0.0010487
-0.0005378
y
g
0.0975889
1 0
9
0.1281797
-0.0481860
1 1
g
-0.0387774
0.0402849
1 p
9
0.0132223
13
9
0.0203827
0.0457820
14
9
0.0371 609
0.01 79332
15
9
0.0134441
0.0410187
16
9
-0,0165750
-0.0509825
17
9
-0.0032015
-0.0343246
18
9
-0.0135216
0.0192446
19
9
0.0017566
0.0086593
20
9
0.0228121
0.0072350
£. 1
9
0 0173205
-0 0093913
22
9
0.0125106-
-0.0094816
23
9
-0.0040274
-0.0103916
OA
g
o oniRQ40
-f) 001410?
25
g
—0.0060297
O.0O98442
26
9
0.0025191 -
-0.0006592
9
o 0001 qqo
n oopi^sfi
28
9
0.0029414-
-0.0030550
29
9
-0.0016394
0.0024790
30
9
0,0001 665
-0.0039351
31
9
-0.0038941
-0.0017016
32
9
0.0019461
0.0007681
33
9
-0.0002939
0.0018691
34
9
0.0012660
0.0015093
35
9
0.0010826
-0.0020731
ir
Ov
9
n oonn 70
— U.UuUO 1 fu
10
10
0.0945596-
-0.0201041
11
10
-0.0520582
-0.0176126
12
10
V-f . U O luf Uu
13
10
0.0433028
-0.0380383
14
10
0.0369953-
-0.0027966
15
10
o nifinm ?
U. V 1 ^ Uir- P C-
16
10
—0.0104171
0.0066056
17
10
0.0021050
0.0201193
18
10
0.0090063
-0.0108619
19
10
-0.0353538
-0.0026556
20
10
-0.0224201 -
-0.0080926
21
10
0.0036543
0.0018356
22
10
0.0050062
0.0203830
23
10
0.0199758
-0,0037585
24
10
0 0092954
25
10
0.0056907-
-0.0044669
26
10
-0.0048370
0.0016645
27
10
-^o oorhoo
l/FUUOJ 1 UVJ
0 00R01 1?
28
10
— 0.0072826
0.0012046
29
10
0.0000602
0.0060144
30
10
o ooipp-rs-
31
10
0.0025178-
-0.0037447
32
10
0.0008308-
-0.0019507
33
10
0.0002341
-0.0009449
34
10
-0.0014879-
-0.0000093
35
10
-0.0014313
-0.0008354
36
10
-0.0003570
0.0004671
11
11
0.0543322-
-0.0547288
12
11
0.0054143-
-0.0095228
13
11
—0 0401906
0 005501*5
14
11
0.0080835-
-0.0413614
15
11
0.0017171
0.0289322
16
11
o 01401 ^fi-
-0 00K41fifi
17
11
— 0.0171 108
0.0175019
18
11
—0.0127989
-0.0005971
19
1 1
0 01 147 4fi
20
11
n 01 1 3787-
-0 0239305
21
11
0.0092806-
-0.0367834
22
11
-0.0093740-
-0.0163775
23
11
0.0038490
0.0136794
24
11
0.0127396
0.0121180
25
11
0.0055793-
-0.0012766
26
11
0.0032045
0,0050467
27
11
-0.0011883-
-0.0030951
28
— fl 000RP14-
_n ooor?qo
\J . PJ UuO£ 3 PJ
29
1 1
—0 0093226
0 0004548
30
11
—0.0016207
0.0050752
31
11
— 0 0G1fifi11
o oorr^pi
32
11
—0.0024010-
-0.0006463
33
11
0.0055568-
-0.0004803
34
1 1
0.0012406-
-0 ooirip^
35
1 1
0.0006785 -
-0.0036583
36
11
0.0005880-
-0.0007308
12
12
-0.0035280-
-0+01 17964
13
12
-0.0280059
0.0864102
14
12
0,0089681 -
-0.0320668
15
12
-0.0283317
0.0124872
16
12
0.0208803
0.0057370
17
12
0.0342734
0.01 72570
18
12
1 9
12
0 0032037
0 0043292
20
12
-0.0040581
0.0172980
21
12
22
12
0 0074177 -
-0 007R4?fi
23
12
0.021 5777 -
-0.0166812
24
12
0 01P1406-
_n ooqm
25
12
—0 00S^4.?5
o m 10100
26
12
-0.0196457
0.0054338
27
12
-0.0004228-
=0.0017182
28
12
0,0004024
0.0024269
29
12
-0.0008339-
-0.0049254
30
12
0.0037915-
-0.0034240
31
12
0.0001491
0.0046159
32
12
-0.0017118
0.0041940
33
12
0 0041PRR
34
12
n nop4R7R
35
23
0 0014841 -
-0.0020619
36
12
o nonpiRp
-U.UU 1 uUOo
13
13
0 OKI -=;4£l1
U.UDatDD I
14
13
0 0315333
0_0446234
1 5
13
— U.UZo I UD I -
-u.uuH-yo.cy
1 o
IO
U.UUUD 1 0*
17
13
0. 0201 122
18
13
— U.UUDOQ ID-
-U.UOD I DO I
1 Q
i y
1 <D
— u.uuDutsyH -
20
13
0.0048913
21
13
-0.0181694
0.0115969
22
13
-0.01 69455
0.0178453
23
13
-0.0104578 =
-0.00751 12
24
13
-0.0036235-
-0.0003824
25
13
0.0073795-
-0.0151883
26
13
0.0027230
0.0014151
27
13
-0.0059813-
-0.0041287
28
13
0.0000983
0.0035308
29
13
-0.0011458-
-00019753
30
13
0,0146742-
-0.0000192
31
13
0.0056871
0.0013250
32
13
0.0072556
0.0002229
33
13
0.0036684
0.0067767
34
13
-0.0080803
0.0012768
35
13
-0.0011849
0.0044812
36
13
0.0007685
0.0037948
14
14
-0.0505657-
-0.0063741
15
14
0.0061707-
-0.0256132
Table 4.32.1, continued
GEM-T1 Normalized Coefficients (x 106)
Value Index Value Index Value
n
c
S
n
m
C
S
n
m
C
S
10
14
-0.0191226
-0.0382895
17
14
-0.0133370
0.0117613
18
14
-0.0092828-
-0.0109400
14
-0.0051227
-0.0126448
20
14
0.0103228-
-0.0117620
21
14
0.0187760
0.0086994
14
0.0087260
0.0102407
23
14
0.0046108-
-0.0032737
24
14
-0.0186436
0.0014570
25
14
-0.0219418
0.0132058
26
14
0.0039290
0.0056350
27
14
0.0119702
0.0066379
28
14
-0.0021064
-0,0065025
29
14
-0.0051525
0.0019409
30
14
-0.0000327-
-0.0025522
ol
14
-0,0072828
0,0012481
32
14
0.0046569
0,0069216
33
14
0.0092319
0.0025099
OH
14
-0.0010431
-0.0002878
35
14
-0.0004836-
-0.0001238
36
14
-0.0048384-
-0.0040665
To
15
-0.0180948
-0,0080854
16
15
-0.0125321
-0.0322958
17
15
0.0049435
0.0057493
18
15
-0.0377619
-0.0198247
19
15
-0.0183164-
-0.0127675
20
15
-0.0227306-
-0.0004135
21
15
0.0166205
0.0149837
22
15
0.0279373
0.0031033
23
15
0.0177318-
-0.0022813
OA
15
0,0098097
-0.0135286
25
15
-0.0019899
-0.0022710
26
15
-0.0113797
0.0047010
£.1
15
—0.0043373
0.0001002
28
15
-0.0082106
0.0053751
29
15
-0,0012718-
-0.0024918
•jyJ
15
0,0028146
-0.0092780
31
15
0.0004541
-0.0043732
32
15
0.0039107-
-0.0049375
33
15
-0.0030055
0.0021726
34
15
0.0007643
0.0030005
35
15
0.0002609
0.0028304
36
15
-0.0018223
0.0018634
16
16
-0.0324114
-0.0043686
17
16
-0.0290683
0.0018848
1 R
16
0.0097880
0.0050024
19
16
-0.0199047
-0,0119326
20
16
-0.0106685
0.0016919
0^
e l
1 6
0.0087331
-0.0051553
22
16
0.0000892-
-0.0049265
23
16
0.0049029
0.0117671
24
16
—0.0004908
0.0062766
25
16
0.0030419
-0.0127972
26
16
0.0058241 -
-0,0041510
27
16
0.0065849
-0.0041072
28
16
-0.0083073-
-0.0076860
29
16
-0.0021980-
-0.0055327
30
16
0.0006189
0.0056597
31
16
-0.0045141
0.0048013
32
16
0.0029187
0.0041010
33
1 6
—0.0003879
0.0019724
34
16
0.0011337
-0.0026578
35
16
0.0001314-
-0.0013369
36
16
0.0013405
-0.0020055
17
17
-0.0383106-
-0.0206234
18
17
0.0061142
0.0087663
19
17
0.0279459
-0.0108837
20
17
0.0042934
-0.0089776
21
17
-0.0067459
0.0008396
22
17
0.0138079
-0.0111258
23
17
-0.0072125
-0.0066031
24
17
-0.0084625
0.0018817
25
17
-0.0083083
0.0005500
26
17
-0.0048890
0.0082820
27
17
0.0055505
0.0015914
28
17
0 0045201
-0.0042606
29
17
0.0045708
-0,0027849
30
17
0.0010057-
-0.0015438
31
17
—0.0059335
0.0025686
32
17
-0.0036602
0.0018909
33
17
-0.0021426
0.0030432
34
17
0 0025764
35
17
0.0033863
-0.0024391
36
17
0.0021679-
-0.0008490
18
18
-0.0044492
-0.0050647
19
18
0.0216467-
-0.0031131
20
18
0.0105771
0.0013024
21
18
0.0168304
-0.0065691
22
18
0.0070311
-0.0102955
23
18
-0.0019056
-0.0063023
24
18
n 004107?
-0. 0050725
25
18
-0.0013004-
-0.0106697
26
18
-0.0090148
0.0075516
27
18
—0 O0R1RQR
0 00^0(194
28
18
0.0003645
-0.0008295
29
18
-0.0020095
-0.0001378
30
18
fl 000101=1
U .UUUvJUiJ J
o oonftsio
31
18
0,0025645
0.0008491
32
18
0.0022105-
-0.0014795
33
18
0.0008702
-0.0012393
34
18
-0.0027232
-0.0000869
35
18
0.0015003
-0.0005937
36
18
0.0000693
0.0005306
19
19
0.0064638
0.0104244
20
19
-0.0070980
0.0084586
21
19
—0 020951 5
0 01 5R7Q0
22
19
0.0066210-
-0.0046952
23
19
-0,0086827
0.0074916
24
19
0 000^374
-0 01 ^017Q
25
19
0.0091802
0.0021317
26
19
0.0016260
0.0007199
27
19
n OO0Q?^4
-0 nofipio^
28
19
0.0044182
0.0138003
29
19
-0.0021920
0.0015059
30
19
-0.0056266
-0.0026384
31
19
0.0020172
0.0028162
32
19
0.0030881-
-0.0015146
33
19
0.001 6796
0.0002369
34
19
0.0005953
-0.0010032
35
19
-0.0029031
0.0005932
36
19
—0 ooopfifip
-0 OOOORQ^I
20
20
0.0017085
-0,0135051
21
20
-0.0190411
0.0185361
22
20
0 01 477AQ
23
20
0.0172248
-0.0090475
24
20
-0.0060619
-0.0003298
25
20
n nnRR? 1 7
U.uVDO^ 1 /
26
20
0.0094830
-0.0109488
27
20
0.0029525
0.0030106
28
20
-0.0009805
0.0011271
29
20
-0.0047996
0.0031625
30
20
-0.0000276
0.0035666
31
20
0.0018947
0.0005595
32
20
-0.0015881
0.0014199
33
20
0.0020822
-0.0007525
34
20
0.0008917
-0.0003975
35
20
-0.0007745
-0.0008763
36
20
-0.0009128
-0.0008661
21
21
0.0024775
-0.0068510
22
21
-0.01 32244
0.0075983
23
21
0.0108195
0.0076428
24
21
0.0105744
0.0011197
25
21
0.0053968
0.0031404
26
21
-0.0003907
-0.0024147
27
21
0.0020712
-0.0045432
28
21
0.0024900
0.0002697
29
21
-0.0093986
-0.0059494
30
21
-0.0074518
-0.0030982
31
21
0.0023358
0.0035767
32
21
0.0011690
0.0058848
232
EXPLANATORY SUPPLEMENT
Table 4.32.1, continued
GEM-T1 Normalized Coefficients (x 106)
Index Value Index Value Index Value
n
m
C
S
n
m
C
S
n
m
C
S
33
21
0.0007319
-0.0008394
34
21
0.0013743
-0.0006590
35
21
0.0013843
0.0027357
36
OH
£A
U,UUUr4l I
a Anon 1 7c
oo
oo
n nm /icoo
— u.uu I *fojio
n AA.H7H QO
oo
oo
d-d.
n aao 1 a a c
24
22
-0.0017322
-0.0013033
25
22
-0.0018741
-0.0017578
26
22
0.0109119
0.0091610
27
22
-0.0001430
0.0029341
28
22
-0.0048465
0.0005118
29
22
0.0096562
0.0044395
30
22
0.0032458
-0.0055410
31
22
-0.0062664
-0.0057325
32
22
-0.0046595
0.0007455
33
22
-0.0040992
-0.0010974
34
22
0.0008294
0.0004221
35
22
0.0000159
0.0033003
36
22
U.UUUO ( OtL
U.00UDOO4
23
23
U.U00044O
O.UUUdUoO
24
do
A ArtOH AtK
-O.UUyUU55
25
23
0.0045743-
-0.0024633
26
23
0.0023537
0.0089516
27
23
-0.0053881 -
-0.0027372
28
23
—0.0026403
0.0063891
29
23
—0.00501 18
-0,0000780
30
23
-0.0015833-
-0,0053377
31
23
0.0095047
0.0056544
32
23
0.0038573
0.0004759
33
23
—0.0006277-
-0.0043975
34
23
-0.0009300-
-0.0021403
35
23
-0.0023783
-0.0015551
36
23
-0.0012168-
-0.0005S18
24
24
0.0023438-
-0.0012129
25
24
0.0036065-
-0.0036584
26
24
—0.0013736
0.0121837
27
24
-0,0019361
0.0026238
28
24
0.0068762-
-0.0150841
29
24
-0.0025342
0.0037024
30
24
-0.0025120-
-0.0000377
31
24
-0.0038298-
-0.0019101
32
24
—0.0065905
0.0053251
33
24
0.0039903-
-0.0004821
34
24
0.0067267
0.000851 1
35
24
0.0025277
0.0021953
36
24
0.0006589-
-0.0014201
25
25
0.0049455
0.0040141
26
25
-0.0039876
0.0082488
27
25
0.01 18014
0.0031453
28
25
0.001 1295 -
-0.0048096
29
25
0,0083057
0.0036286
30
25
0,0088475
-0.0056062
31
25
-0.0077449-
-0.0002757
32
25
-0.0131730
0.0077043
33
25
—0.0012628-
-0.0048224
34
25
0.006251 6-
-0.0082080
35
25
— 0.0033892
0.0015231
36
25
0.0000270
0.0086477
26
26
0.0034281 -
-0.0042690
27
26
-0.0050080
0.0040035
28
26
0.0034421
0.0016811
29
26
0.0062584-
-0.0036878
30
26
-0.0032087
0.0081399
31
26
—0.0046923
0.0003639
32
26
—0.001051 1 -
-0.0015337
33
26
n Arto H o H o
0.0081343
U.Uuobjob
34
26
0 0010R1B-
\J.\A\J &\J\J\J P
35
26
—0 01 4343R-
-0 0001412
36
26
0.0084469
0.01 10849
27
27
0.0068945
0.0034538
28
27
—0.0099247
0.001 3336
29
27
-0.0074725-
-0.0021507
30
27
—0.0019206
0 007R1SP
31
27
o nofi^qfio
VbUyU^V uy
0.0122525
32
27
-0.0030949-
-0.0030221
33
27
-0.0103323-
-0.0021297
34
27
0.0068974-
-0.0005442
35
27
0.0027787-
-0.0191290
36
27
—0.0101912
0.0042305
28
28
0.0067689
0.0019493
29
28
0.01 03226-
-0.0019625
30
28
-0.0089658-
-0 0051 589
31
28
0 0003428
0 0017261
32
28
0.0015717
0.0023530
33
28
— 0.0108851
34
28
-0.0081 979
35
28
—0.01 08936-
-0.0233542
36
28
0.0069480
0.0056619
29
29
0 0086336
0.0031601
30
29
0.0048227
0.0001240
31
29
-0.0054215-
-O0O59502
32
29
-0.0033428
0.0025024
33
29
-0.0213209
0.0001932
OA
29
—0.00381 13-
-0.0044842
35
29
—0.0039791
0.0005664
36
29
— 0.0013415-
30
30
-0.0015075-
-0.0004221
31
30
—0.0024527
0 0084230
32
30
0.0082893
0.0016713
33
30
0.0025217-
-0.0134935
34
30
-0.0061161
0.0000268
35
30
0.0037128-
0.0028852
36
30
-0.0015551-
-0.0020261
31
31
-0.0002319-
-0.0000737
32
31
-0.0007616-
-0.0027125
33
31
0.0001 745
0.0011566
34
31
0.0023534
0.0022983
35
31
0.0013305
0.0010339
36
31
-0.0035516
0.0013604
32
32
0.0007277
0.0005068
33
32
0.0024100-
0,0000219
34
32
-0.0008166-
-0.0011610
35
32
-0.0039791
0.0005664
36
32
-0.0008804-
0,0003000
33
33
-0.0002676-
-0.0003603
34
33
0.0010234
0.0013813
35
33
-0.0002881
0.001 6655
36
33
-0,0023972-
-0.0026170
34
34
-0.0002093-
-0.0006039
35
34
-0.0005168
0,0001237
36
34
0.0008324
0.0017560
35
35
0.0000938-
-0.0001450
36
35
0.0002152-
0.0007347
36
36
0.0001484
0.0004197
* Cgo does not include 1he zero-frequency term; see Equation 4,332-5 for the adjusted value.
" Cgi and Sji should be the IERS values; see Equations 4,32-3 and 4.32-4 for recommended values.
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 233
should be added to the geopotential model (Lambeck, 1971). This gives normalized
coefficients of
C21(IERS) = -0. 17 x 1CT8, (4.32-3)
52i(IERS) = 1. 19 X 1CT9. (4.32-4)
For consistency with the IERS Terrestrial Reference Frame, the C2KIERS) and
521(IERS) should be used in place of C2l (GEM-T1) and 52i(GEM-Tl).
4.33 Solid Earth Tides
The solid Earth tide model is based on an abbreviated form of the Wahr model
(Wahr, 1981) using the Earth model 1066A of Gilbert and Dziewonski (1975). The
Love numbers for the induced free space potential, k, and for the vertical and
horizontal displacements, h and /, have been taken from Wahr's thesis, Tables 13
and 16. The long period, diurnal, and semi-diurnal terms are included. Third-degree
terms are neglected.
4.331 Calculations of the Potential Coefficients The solid-tide-induced free-space
potential is most easily modeled as variations in the standard geopotential coef-
ficients Cnm and Snm (Eanes et al, 1983). The Wahr model (or any other having
frequency-dependent Love numbers) is most efficiently computed in two steps. The
first step uses a frequency-independent Love number k2 and an evaluation of the
tidal potential in the time domain from a lunar and solar ephemeris. The second
step corrects those arguments of a harmonic expansion of the tide-generating po-
tential for which the error from using the k2 of step 1 is above some cutoff.
The changes in normalized second-degree geopotential coefficients for step 1
are
v 3 j=2 j
5 GM <-~> ri
j=2 1
AC22 - ,'A522 = 1 Mh^t, E ^2(sin 4>j)e^,
12 V 5 GM ^— ' ri
j=2 J
where
(4.331-1)
(4.331-2)
(4.331-3)
#2 = nominal second degree Love number;
-Re = equatorial radius of the Earth;
234 EXPLANATORY SUPPLEMENT
GM = gravitational parameter for the Earth;
GMj = gravitational parameter for the Moon (j = 2) and Sun (j = 3);
Tj = distance from geocenter to Moon or Sun;
<j»j = body fixed geocentric latitude of Moon or Sun;
\j - body fixed east longitude (from Greenwich) of Sun or Moon.
The changes in normalized coefficients from step 2 are:
ACnm-iASnm=Am TSkJlJ M n + m ™* (4.331-4)
' \ -i ) n + m odd '
v(n,»i)
where
A - M)
f 1 if m = 0
fc.= (0 ifmyo ; (4.331-5)
6ks - difference between Wahr model for k at frequency s and the nominal
value k% in the sense ks — &2;
Hs - amplitude (m) of term at frequency s from the Cartwright and Taylor
(1971) and Cartwright and Edden (1973) harmonic expansion of the
tide-generating potential;
n = six vector of multipliers of the Doodson (1921) variables;
0 = the Doodson variables;
6S20 — 0.
The Doodson variables are related to the fundamental arguments of the nuta-
tion series (see Chapter 3) by
s = F+ f2 = (Moon's mean longitude);
h = s — D = 03 (Sun's mean longitude);
P = s - I = 04 (longitude of Moon's mean perigee);
W = ~S2 = 0s (negative longitude of Moon's mean node);
P\ = s — D — I' = 06 (longitude of Sun's mean perigee);
T = 9g + -n-s = 0\ (time angle in lunar days reckoned from lower transit);
By = mean sidereal time of the conventional zero meridian.
The normalized geopotential coefficients (C,„„,Sn,„) are related to the unnormal-
ized coefficients (Cnm,Snm) by
Cim = N„mCnm, (4.331-6)
Snm=NnmSnm, (4.331-7)
Nm
(n + m)\(2n + \)(2-t>Qmyl
(« + m)\
(4.331-8)
TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 235
Table 4.331.1
Step 2 Solid Tide Corrections When k2 = 0. 30 in Step 1 (Using a
Cutoff Amplitude of 9 x 10~12 iorAm6ksHs)
Long-Period Tides {n = 2, m = 0)
None except zero-frequency tide.
Diurnal Tides (n = 2, m = 1 )
n, argument multipliers
Doodson Number t
s
h
p
N'
P\ An
SksHsX. 1012>
145.555
(Oi) 1
0
0
0
0
-16.4
163.555
(Pi) 1
-2
0
0
0
-49.6
165.545
0
0
-1
0
-9.4
165.555
(/Ci) 1
0
0
0
0
507.4
165.565
0
0
1
0
73.5
166.554
M) 1
1
0
0
-1
-15.2
Semi-Diurnal Tides (n = 2, m - 2)
ri, argumen) multipliers
, »— -
Doodson Number r s ft p Af p\ Amf>ksHs x 1012
255.555 {M2) 2 0 0 0 0 0 39.5
273.555 (Sj) 2 2 -2 0 0 0 18.4
Using a nominal £2 of 0.3 and an amplitude cutoff of 9 x 10~12 change in normalized
geopotential coefficients, the summation S(n,m) requires six terms for the diurnal
species (n = 2, m = 1) modifying C21 and S2i and two semidiurnal terms (n - 2,
m = 2) modifying C22 and S22- With the exception of the zero-frequency tide, no long-
period terms are necessary. Table 4.331.1 gives required quantities for correcting
the (2,1) and (2,2) coefficients. The correction to C2o is discussed in more detail
below.
Example: The step 2 correction due to the K\ constituent is
(AC21 x ]012)a-, = 507.4sin(r + .f)
= 507.45^(9,, + tt)
= -507.4 sin(8(), (4.331-9)
(AS21 x 10l2k, = -507.4cos(e^). (4.331-10)
The total variation in geopotential coefficients due to the solid tide is obtained
by adding the results of step 2 (Equation 4.331-4) to those of step 1 (Equations
4-331-l, 4.331-2, and 4.331-3).
236
EXPLANATORY SUPPLEMENT
4.332 Treatment of the Permanent Tide The mean value of AC20 from equation
4.331-1 is not zero, and this permanent tide deserves special attention. Ideally, the
mean value of the correction should be included in the adopted value of C2o and
hence not included in the AC20. The practical situation is not so clear, because
satellite-derived values of C2n, as in the GEM geopotentials, have been obtained
using a mixture of methods, some applying the corrections and others not applying
it. There is no way to ensure consistency in this regard short of re-estimating C20
with a consistent technique. If this is done, the inclusion of the zero frequency term
in Equation 4.331-1 should be avoided, because k2 is not the appropriate Love
number to use for such a term.
The zero frequency change in C2q can be removed by computing AC20 as
ACjo = AC20(Equation 4.331-1) - (AC20) , (4.332-1)
where
<AC20) = A0HQk2
-(4.4228 x Kr8)(-0.31455)/t2
= -1.39119 x KT8Jt2.
Using k2 = 0.30 then
<AC20) = -4. 1736 x 10~9, (4.332-2)
or
(AJ2) = - (AC20) = 9. 3324 x 10~9. (4.332-3)
The decision to remove or not to remove the mean from the corrections depends
on whether the adopted C20 does or does not already contain it and on whether k2 is
a potential "solve for" parameter. If k2 is to be estimated then it must not multiply
the zero frequency term in the correction. In the most recent data reductions leading
to GEM-T1, the total tide correction was applied. If we assume the more recent
data has most of the weight in the determination of C20 then we conclude that the
permanent deformation is not included in the GEM-T1 value of C2Q. Hence, if k2 is
to be estimated, first (AC2o) must be added to C20 and then AC20 should be used
in place of C20 of Equation 4.331-1. The k2 used for restoring the permanent tide
should match what was used in deriving the adopted value of C2o-
The GEM-T1 value of C20 is -484. 16499 x 10 6 and does not include the
permanent deformation. The tidal corrections employed in the computations leading
to GEM-Tl were equivalent to Equation 4.331 1 with k2 = 0.30. Let C2o denote
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 237
the coefficient that includes the zero-frequency term; then the GEM-Tl values of
-q with the permanent tide restored are
C*2n{GEM -T\) = -484. 16499 x 10~6 - (1. 391 19 x 10"8) x 0. 30. (4.332^4)
C*m(GEM -71) a -484. 169025 x 10~6. (4.332-5)
These values for C*2fj are recommended for use with the respective gravity field and
should be added to the periodic tidal correction given as AC2q in Equation 4.332 1
to get the total time-dependent value of C2o-
4.333 Solid Tide Effect on Station Coordinates The variations of station coordi-
nates caused by solid Earth tides predicted using Wahr's theory are also most
efficiently implemented using a two-step procedure. Only the second degree tides
are necessary to retain 0.01 m precision. Also terms proportional to y, h+, h—, z, /+,
u'+, and w— are ignored. The first step uses frequency-independent Love and Shida
numbers and a computation of the tidal potential in the time domain. A convenient
formulation of the displacement is given in the documentation for the GEODYN
program (Martin et ai, 1976, p. 5.5-1) . The vector displacement of the station due
to tidal deformation for step 1 can be computed from
where
GMjr4
GMRj
[3/2(Rj ■ r)]R; +
i hy , \ _ 3 hi
(4.333-1)
GMj = gravitational parameter for the Moon (j - 2) or the Sun (/ = 3);
GM = = gravitational parameter for the Earth;
R/, Rj = unit vector from the geocenter to Moon or Sun and the magnitude of
that vector;
r, r = unit vector from the geocenter to the station and the magnitude of that
vector;
h2 = nominal second-degree Love number;
l2 = nominal Shida number.
If nominal values for h2 and l2 of 0.6090 and 0.0852, respectively, are used with
a cutoff of 0.005 m of radial displacement, only one term needs to be corrected in
step 2. This is the K\ frequency where h from Wahr's theory is 0.5203, Only the
radial displacement needs to be corrected and to sufficient accuracy this can be
implemented as a periodic change in station height given by
6hSTA = ShKl HK, { - \/ ^- ] 3 sin <f> cos <A sin(©^, + A), (4.333-2)
238 EXPLANATORY SUPPLEMENT
where
%i (Wahr) -/i2 (Nominal) = -0.0887;
f/^ — amplitude of K\ term (165.555) in the harmonic expansion of the tide
generating potential = 0. 36878 m;
cf> = geocentric latitude of station;
A = east longitude of station;
©Ki = Ki tide argument = t + s = Qg + n,
or, simplifying,
ShsTA = -0. 0253 sin <$> cos <f> sin(6^ + A). (4.333-3)
The effect is maximum at </> = 45° where the amplitude is 0.013 m.
There is also a zero- frequency station displacement that may or may not be
included in the nominal station coordinates. The mean correction could be removed
analogously to the discussion above. When baselines or coordinates are compared at
the few-cm level, care must be taken that the correction was handled consistently.
It is essential that published station coordinates identify how the zero-frequency
contribution was included.
If nominal Love and Shida numbers of 0.6090 and 0.0852, respectively, are used
with Equation 4.333-1, the permanent deformation introduced is
Ar r = i/— (0.6090)( -0.31455) ( -sin2^- - J (4.333^)
V 47r \2 2 J
-0. 12083 ( I sin2 4> ~ I: ) m (4.333-5)
2 " 2
in the radial direction, and
Ar ' eP = \/^-(°- 0852)(-0. 3 1455)3 cos <p sin 0
V 47T
= -0. 0507 1 cos 4> sin 4> m (4.333-6)
in the north direction.
4.334 Rotational Deformation Due to Polar Motion The variation of station coor-
dinates caused by the polar tide should be taken into account. Let us choose x, f,
and z as a terrestrial system of reference. The z-axis is oriented along the Earth's
mean rotation axis, the x-axis is in the direction of the adopted origin of longitude,
and the y-axis is oriented along the 90° E meridian.
The centrifugal potential caused by the Earth's rotation is
V = 1 /2[r2|0|2 - (r- SI)2],
(4.334-1)
4 / -TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 239
kere ^ = Q{m\X + m2y + {\+m-i)z). H is the mean angular velocity of rotation of the
Earth; mt are small dimensionless parameters, mi, mj describing polar motion, and
0^ describing variation in the rotation rate; r is the radial distance to the station.
Neglecting the variations in m-s that induce displacements that are below the
mm level, the m\ and m2 terms give a first-order perturbation in the potential V
(Wahr, 1985)
AV(r, 9, A) = — sin 20(^1 cos A + m2 sin A), (4,334-2)
where 0 is the colatitude, and A is the eastward longitude.
Let us define the radial displacement Sr, the horizontal displacements S® and
5A positive upward, south, and east, respectively, in a horizon system at the station
due to AV using the formulation of tidal Love numbers (Munk and MacDonald,
1960, pp. 24-25).
AV
Sr - h— , (4.334-3)
g
Se = --d&AV, (4.334^)
g
Sx = -^-r8x^V, (4.334-5)
g sin B
where g is the gravitational acceleration at the Earth's surface, and h, I are the
second-order body tide displacement Love numbers.
In general, these computed displacements have a nonzero average over any
given timespan because mi and m%, used to find AV, have a nonzero average. Con-
sequently, the use of these results will lead to a change in the estimated mean
station coordinates. When mean coordinates produced by different users are com-
pared at the centimeter level, it is important to ensure that this effect has been
handled consistently. It is recommended that mi and m2 used in Equation 4.334-2
be replaced by parameters defined to be zero for the terrestrial reference frame
(ITRF) discussed in the previous section.
Thus, define
xp = m]-x, (4.334-6)
yP=-m2-y, (4.334-7)
where x and y are the values of mi and -m2 for the ITRF. Then, using h = 0. 6,
' = 0.085, and r = a = 6.4 x 106m,
240
EXPLANATORY SUPPLEMENT
Sr - -32 sin 2&(xp cos A - yp sin A) mm, (4.334-8)
Se = -9 cos 2Q(xp cos A - yp sin A) mm, (4.334_9)
Sx - 9 cos Q(xp sin A + yp cos A) mm. (4,334-10)
for xp and yp in seconds of arc.
Taking into account that xp and yp vary, at most. 0.8 arcsec, the maximum radial
displacement is approximately 25 mm, and the maximum horizontal displacement
is about 7 mm.
If X, Y, and Z are cartesian coordinates of a station in a right-hand equatorial
coordinate system, we have the displacements of coordinates
[dX, dY, dZ\T = RT[Se,Sx,Sr]T, (4.334-1 1)
where
R
cos (-) cos A cos 6 sin A - sin 0
- sin A cos A 0
sin 0 cos A sin 9 sin A cos 0
(4.334-12)
Equation 4.334 10 can be used for determination of the corrections to station
coordinates due to polar tide.
The deformation caused by the polar tide also leads to time-dependent pertur-
bations in the C2\ and S2] geopotential coefficients. The change in the external po-
tential caused by this deformation is kAV, where AV is given by Equation 4.334-2,
and k is the degree-2 potential Love number. Using k = 0. 30 gives
C2] = -1.3 x W-g(xp), (4.334-13)
§21 =-1.3 X lO-g(-yp), (4.334-14)
where xp and yp are in seconds of arc and are used instead of m\ and -m2 so that
no mean is introduced into C2\ and S2\ when making this correction.
4.34 Ocean Tide Model
The dynamical effect of ocean tides is most easily implemented as periodic variations
in the normalized geopotential coefficients. The variations can be written as follows
(Eanes et ai, 1983):
AC,„„ - iASnm = FfUu £ £<f£- T iSLy*' , (4.34- 1 )
s{n,m) +
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 241
where
AirGpw
nm '
8
(n +m)\
(rt-m)!(2«+l)(2-5om)
1 2
l +k
277T< (434-2>
where
8 = GEIa2 = 9.798261ms-2;
G - the universal gravitational constant = 6. 673 X 10_n m kg~' s-2;
pw = density of seawater = 1025 kgm-3;
k'n = load deformation coefficients (k'2 = -0. 3075, k'3 = -0. 1954, k'4 = -0. 132,
k'5 = -0. 1032, k'h = -0.0892);
C* , 5t„ = ocean-tide coefficients in meters for the tide constituent s (see Ta-
ble 4.34.1);
Ss = argument of the tide constituent s as defined in the solid tide model (see
previous subsection).
The summation, , implies addition of the expression using the top signs
(the prograde waves C*snm and S$„m) to that using the bottom signs (the retrograde
waves C~m and S~,„). The ocean-tide coefficients C%„m and S$nm as used here are
related to the Schwiderski (1983) ocean-tide amplitude and phase by
Clm - /S*m = WC7n„,<,''(f""*'+x'\ (4.34-3)
snm sunt snni 1 > '
where
Cjnm = ocean-tide amplitude for constituent s using the Schwiderski notation;
e^, = ocean-tide phase for constituent s;
(0 "I Semidiurnal, and long period
7r / 2 > K\ ; (or in general for constituents with Hs > 0),
— 7r 12 ) 0\, P\, Q\ \ (or in general for constituents with Hs < 0).
For clarity. Equation 4.34-1 is rewritten in two forms below:
or
or
AC„m = Fnm Y, [(d. + C-J cos 0, + CSS* + S-J sin ©.,], (4.34-4)
ACnm = Fnm Y [$L sin(0J + e+mm + Xs) + C~m sin(©1 + e;„„, + Xs)l (4.34-5)
i(«,m)
A5nffl = Fm„ [CSLi + S»„J co* 0. - (C, + C" „) sin &„}, (4.34-6)
AS,™ = Fnm Y [Qm cos(0.t + e^, + Xs)
s(n,m)
+ C~m cos(0J + e-,„ + x,)l (4-34-7)
242
EXPLANATORY SUPPLEMENT
Table 4.34.1
Ocean Tide Coefficients from the Schwiderski Model
Argument
n
m
-
^snm
^snm
Number
(cm)
(dea)
vucy/
(cm)
(cm)
057. 555
Ssa
2
0
.6215
221.672
-.8264
.9284
057. 555
Ssa
3
0
.0311
1.735
.0019
.0621
057. 555
Ssa
4
0
.1624
92.674
.3244
-.0152
057. 555
Ssa
5
0
.2628
251.737
-.4991
-.1647
057. 555
Ssa
6
0
.4363
145.744
.4912
-.7213
065. 455
Mm
2
0
.5313
256.900
-1.0428
-.2046
065. 455
Mm
3
0
.0317
94.298
.0632
-.0047
065. 455
Mm
4
0
.0998
69.054
.1863
.0713
065. 455
Mm
5
0
.2279
292.291
.4218
.1729
IVlm
et
.\JOU\J
ftft?
nRd7
mid
075. 555
Mf
2
0
.8525
251.956
-1.6211
-.5281
075. 555
Mf
3
0
.0951
148.236
.1001
-.1617
075. 555
Mj
4
0
.2984
102.723
.5822
-.1315
075.555
Mf
5
0
.2960
223.167
-.4050
-.4318
075.555
M}
6
0
.0880
107.916
.1675
-.0542
The summation over s(n, m) should include all constituents for which Schwiderski
has computed a model. Except for cases of near resonance, the retrograde terms
do not produce long-period (> 1 day) orbit perturbations for the diurnal and semi-
diurnal tides. The root mean square of the along-track perturbations on LAGEOS
due to the combination of all of the retrograde waves is less than ± 5 cm.
For computing inclination and node perturbations, only the even-degree terms
are required, but for the eccentricity and periapsis the odd-degree terms are not
negligible. Long-period perturbations are only produced when the degree {«) is
greater than 1 and the order (m) is 0 for long-period tides, 1 for diurnal tides, and 2
for semidiurnal tides. Finally, the ocean-tide amplitudes and their effect on satellite
orbits decrease with increasing degree, so truncation above degree 6 is justified for
LAGEOS.
Thus, for the diurnal tides {Q\ , 0\ , P\ , K\ ) only the n = 2, 3, 4, 5, 6 and m = 1
terms need be computed. For the semidiurnal tides (N2, M2% S2l K2) only n = 2, 3,
4, 5, 6 and m — 1 terms need be computed. For the long-period tides (Ssa, Mm, Mf)
only n = 2, 3, 4, 5, 6 and m - 0 terms need be computed. Table 4.34.1 gives the
values required for each of the constituents for which Schwiderski has computed
a model. Note that the units in Table 4.34.1 are centimeters and hence must be
scaled to meters for use with the constants given with Equation 4.34-1.
Table 4.34.1, continued
Ocean Tide Coefficients from the Schwiderski Model
Lfgu merit
n
m
6snm
esnm
Number
(cm)
(deg)
(cm)
(cm)
135. 655
O1
2
1
.5373
313.735
-.3715
-.3882
135.655
Q,
3
1
.3136
107.346
,0935
.2994
135.655
Q1
4
1
.2930
288.992
-.0953
-.2770
135.655
Oi
5
1
.2209
112.383
.0841
.2042
135.655
Oi
6
1
.0396
287.824
-.0121
-.0377
145.555
0^
2
1
2.4186
313.716
-1.6715
-1.7481
145.555
Oi
3
1
1.3161
83.599
-.1467
1 .3079
145.555
o,
4
1
1 .4301
276.282
.1565
-1.4215
145.555
5
1
.9505
109.128
.3115
.8980
145.555
Oi
6
1
.1870
282.623
-.0409
-.1825
163.555
Pi
2
1
.9020
313.912
-.6256
-.6498
163.555
Pi
3
1
.2976
39.958
-.2281
.1911
163.555
Pi
4
1
.6346
258.31 1
.1286
-.6215
163.555
Pi
5
1
.4130
104.438
.1030
.4000
163.555
Pi
6
.0583
276.591
-.0067
-.0579
165.555
*1
2
1
2.8158
315.113
1.9950
1 .9872
165.555
*1
3
1
.8925
33,752
.7421
-.4959
165. 555
Ki
4
1
1.9121
254.229
.5197
1.8401
165.555
*1
5
1
1.2111
104.672
-.3068
-1.1716
165. 555
K|
6
1
.1645
281 .867
.0338
.1610
245.655
N2
2
2
.6516
321 .788
-.4030
.5120
245. 655
Nz
3
2
.1084
171 .923
.0152
-.1074
245, 655
N2
4
2
.2137
141.779
.1322
-.1679
245. 655
W2
5
2
.0836
5.034
.0073
.0832
245. 655
N2
6
2
.0674
346.544
-.0157
.0656
255. 555
Mr,
2
2
2.9551
310.553
-2.2453
1.9213
255. 555
M2
3
2
.3610
168.623
.0712
-.3539
255. 555
M2
4
2
1.0066
124.755
.8270
-.5738
255. 555
M2
5
2
.2751
356.561
-.0165
.2746
255. 555
M2
6
2
.4130
329.056
-.2124
.3542
273. 555
Sz
2
2
.9291
314.011
-.6682
.6456
273. 555
S2
3
2
.2633
201.968
-.0985
-.2442
273. 555
S2
4
2
.3716
103.027
.3621
-.0838
273. 555
Sa
5
2
.1365
3.772
.0090
.1362
273. 555
S2
6
2
.1726
280.381
-.1698
.0311
275. 555
K2
2
2
.2593
315.069
-.1832
.1836
275. 555
K2
3
2
.0943
195,007
-.0244
-.0911
275. 555
K2
4
2
.1059
103.521
.1029
-.0247
275.555
K2
5
2
.0382
.411
.0003
.0382
275.555
Kz
6
2
.0467
281.357
-.0458
.0092
(1) The Doodson variable multipliers (n) are coded into the argument number (A) after Doodson (1921) as
A = n, (na + SKng + 5). (n4 + 5)(rt5 + 5)(n6 + 5).
(2) For the long-period tides (m = 0), the value oi C£nm used to compute C*nm and Sjnm was twice that
shown to account for the combined effect of the retrograde and prograde waves.
(3) The spherical harmonic decomposition of Schwiderski's models was computed by C. Goad of Ohio
State University.
EXPLANATORY SUPPLEMENT
The (n = 2,m = 2) term for the S2 argument can be modified to account for the
atmospheric tide using the results of Chapman and Lindzen (1970). The modified
values to be used instead of those in Table 4.34.1 are
Ct, = -0.537 cm, (4.34-8)
S| =0.321 cm. (4.34^9)
For the most precise applications, more than the eleven terms listed in Table 4.34.1
need to be modeled. This can be accomplished by assuming that the ocean tide
admittance varies smoothly with frequency and by using the Schwiderski values as
a guide to the interpolation to other frequencies.
4.35 Site Displacement Due to Ocean and Atmospheric Loading
4.351 Ocean Loading In the past decade, displacements due to ocean loading have
been computed by several authors (Goad, 1980; Agnew, 1983; Scherneck, 1983; Pa-
giatakis et al, 1984; Sato and Hanada, 1985), most of them using methods similar
to that described by Farrell (1972) and based on global representations of the ocean
tides like those by Schwiderski (1980).
In all these models, the locally referenced displacement vector E in the vertical
(radial), N-S and E-W local coordinate system at time t can be computed as a
sum of the contributions of n individual ocean tides
a ( A\ cos(wff + 4>i - 6rj)
E(r) = ]>N Af cos(w,f + ^-(5f) (4.351-1)
« { Afwcos(Lj{t + 4>i-Sf-w)
where is the frequency of the tidal constituents and $t the corresponding astro-
nomical argument. The amplitudes A\, Af , Afw, and the Greenwich phase lags tf ,
5fs, 6fw of each tidal component are determined by the particular model assumed
for the deformation of the Earth.
The site displacements resulting from the ocean tides have been compiled by
C. Goad. Tidal loading radial (height) displacement amplitude and phase values for
laser and VLBI station locations have been computed for the M2, S2, K2,N2, Ot,Ku
PuQu Mf, Mm, and Sm ocean tides using models generated by Schwiderski (1978).
The technique of Goad (1980) that uses integrated G reen's functions was employed.
The load-deformation coefficients used in generating the Green's function integrals
were taken from Farrell (1972).
Table 4.351.1 gives the amplitude and phase values for the eleven main tides at
the selected stations. The phase convention of Schwiderski has been preserved. His
phases for the diurnal tides differ by ± 90° from the standard Doodson definition.
Table 4.351.1
Displacement Due to Ocean Loading (cm in amplitude and degree in phase)
<
I
a
£L
s
<
to
<
X
a.
m ^ tfi CO o ^- CO CO CM CD Is OJ q Ol N cq p CO -3; m ra O) O) t- cm cq CO en 05 oi ^
r~+ c\i c\i c\i co cci K ^ cm d ri d ri n cy ri c\ cd co cd 't t- c\i W ^t-'-^t*
w o Q o tfi o> cm cm co a>a> co^i^ s N s s s NKr^r^r^ r^^Nr^r^ n s s ?i
1 7 T T 1 T T T *"
"itT-i-.--^- tfl it CO t- CM CM CO CO LO IC5 CO O CO IO CO CO (O S CO CO K> S N N N (o «j
ppppp ppp-r-p ppppp ppppp ppppp opppp oppo
d r co c\{ cm co ^ to co cm cq oq p cm q s co Tf q r- ^ s 10 q n cs q co n coco^co
co cd ci n d oi ^ co d n o> cd a> to cri K) s ^ d n cd oi K cd K cd^r^T-^T^ ^r^J^rn
c\i cs CD co 10 u) n co co m *r cm o co n m rt id ^toooco oocccdtj ^ ^ n s
f i ll ^ ^ " T-T-rrr^rrrrr^rr-T-r
CO CM CM CM CM CM t- O CM t- t- i- i- i- t- t-t-t-t-t- r r fe^f :E i-t-^-^t- t- t- -i-
ooppp ppppp ppppp ppppp ppppp ppppp a o o o
CO
X
cm co co cm co to r~ cq cq 10 n q q ^ s n 10 q ^ cm iq cm co t~ ^ tj- cm cm cm cm
^ lo cri co o o o -<a- co ro ^- co n co o o ca co
CM CM CM *- t- i- CM IT CO CO icDt- r O t- CO O
III ^ T T IT ITT IT
CO O) N CO
0} CD CD CO
I
C5 ^ O) S r- T-^-cd™
CO CO 03 CO CO CO CC CO
I I I I I I I I I
S
qpppp ppppa qpqpp qppqp qooo
10
<
X
Cl
<
eg
<
CL
^t^-o^N r> q o) q q cq to ifl to q
^3 CO C(j CO t- W C\i Cfi O S ^ T-^ K Ifl
O^-^-^Tj I PI i- T- t C\| r r
!D i- S 7- N CO CO (O
in* co iii k oi cd as Tt1
to in i- i- i- eg oj
^' cti cd od cd cd cd
i- t- -r* i- i- CM CM
in ifl m- n n
ooooo
O) CM CO CJ CO CO CM CM CM W CM W CO rf ^ CM CM
CO CO CO SD O i- h-
in'coco^V ^ion
I to CD CD CO CO S S
1 7 7 7 7
CO CO CO O N LT) N OJ CM CD OJ O LO N CM CO *>~ i- i- t- tJ- LO
pi CT> i- CO co n CO W CM CO 1- CO
COO O Tf O CO CO CO N CO N
CD CM CO DO
- CO GO CO CD
I I I I
CM CO (O CO r- T~
CO CO CO h- "
I I I I
I I
N CO m CO ^ CO ^ CM N CM S O) N O COtOLTJLOCV
ooooo ooooo o o o o OOOO-i-
co co oo co co
o o o o o o
<
X
CL
CL
<
<
X
Q_
<
CO
<
X
rjjrtrj-iLOCM CM N CD t CO N CO ID LO O) CO CD fl- CM1 N t- CO O) CO t- O) t- CO lO W CM CM CO CO
CO 05 oi ai CO CO CO CO CD CO CD CO d CO CO d cci r-1 cci d W CM CM N CM CO N t- t; -p- <r- CO CO
Lfi CO CO CD I I i-CO COCO't^CO tt^CO^CO ■^^^'m'CO ftTfn^'vt ^^^t^-
I I I I 1
O N S N CM CM ^ O tO LO m CO CO CM CO ^ O CO f- Ifl CO i- i- O t- t- CO CD CO CO CO r- r-
t-ooot- 1-; t- -r- q t- t- t-_ t- cm cm CM co to co to co n co to to to ^ n to q co co co
N CO CO Cp CM CM Is* Cp CO CM O CO CD CO i— CO CO U] CM Lfl N CO tf) N OCOOJ^q O) O) q S
CO CD CO CD Ifl LO -i- CO CD oi r- S CO CO 6 N If) N C\i (D CO ! CD LD CO ^ lO ■ ^' ffl N N ^ *i EE' ES
CO O O O CD CO CO ^ CO t- CO i- CO N CM CO CO CD CO tO lO lO lO CO LO lO CO LO LT) LO LO CO CO
i i i i i i i i i f i T i i I I I i i r i i i
o> a> co -i— i- <d ■•— m cm co co n co to co n co en co cm cm cm fl- « cm to n co cooooo
CO CM C\J CM CM CM — t-t-O O O O O — O — ■>-■>- ■>- ^ -r- CM t- r CJ r r — — ^ T
if) rt O S co
I I I
CO CO CD T-
Tt" I* tO
i i 1 r\i
CM CO O) CO CO
CO CO CO
CM CO
p CO N t-;
CO CO CO CO CO
CO CM CM CM CM CM CM CM CM CM
oo o co co cs
CO CO CO S CO
CM CM CM CM CM
o> O) *»
O CO CO CO CD CO CM O CM CM i- h- 1^ r- CO t- N CO t- CO CM CM CM ^- CM CM LO OJ CD ^ fO CO CO
CD O O CM CM CM CM i- CO CO CO CO V lO ^ CD CO CO N CD q (D CO CO tO tp CO S ^ h N * (P
en
<
X
a
a
<
CO
<
X
Q_
CL
<
<
X
CL
CL
CO
LO
I
O O TT
O CD LO
^ CD CO CM LO
LO r— LO CM
II ^ «
ifl O lO (D i-
a> od o
CO CO Tt TLj Tt
lO lO f N
K t- oi cm' ai
^ ^- CO tl" co
O q co q cm
t— co co co cd
r- co co cj> cb en co co
co d co cm t- t: ril J
Tj-Tr^j-Ttco ^<tco^-^r ^ ^
r t IO O CO CO ^3" CM CO -i- CO Ol CO CM CO CM fl) CO CM i- N N lO (D O to N CO ^ N ^ r-- q q
t^t-'t-'cv^ ^j o co cm cd cd i— o oS cd cdojco'cnco oi crj s co ^' 99 2 ^ 3; S S O 2
i— £D CD CO LO LO N S CO CM CO CO N CM CD CO Ol CO CT) CO QO O O O K O O N CO CO CO CO ^ T
1 1 1 1 7 7 7 7 77 77 77 1 7 7 7 7 1 7 7 1 1 1 1 1 7 T
n to co m to to tj- o "d- m co^co^Tt- cm cd to co n cm lo io lo lo ^ ^ ffl oj r- t~ t: ° 5
t m: fl: C\l CM CM. CM CM r- r; t; t; CM t-; -r^ CM — CM ftl r ■ i- t- CO p r; CO r; CM CM <M r; r
i- ^ tj- o co oi n ^ CD co to to to co cm cm *f lo cm s to "t co h- o q q ■— q o q q q i>
2 £ co n co cd cd oj io ai t- cri id w n dcMoi^fLO cri co ^ lo d cd r- co n ^ l> ^m ^
T CO CO CO N N UD tfi CD CD i— CD <T> I LO CO ^ CO t— t— i— t- i 't CM CM CM CM CO n
i i i i - y"^rf i iiii I I i I I iill ii
N N S S (O CD LO CO CM CO CM CM f- CO 'Cf CO CM CO r- CO CM CM CM p EHS^S^ ^ r3 rv
CO CO CO CO CD Ol fl) (O CO O CM O CM O CM CM CM CM lO CO CM CM CM N CM CM CO 't lO LO LO CM CM
i 3
S -o 5
CM CO ^
o
I 'lg °
w ir <
» co I
5 S z
CO
o
E s
.a tr
DC X
5 I
0- > > LL. >
CD
LU CO
is
co w ^
O CO C 4_ ^
O 01 o
_ o
i> (fl o >■ 2 P a &
_ CQ Q_
O O O O -i cn
O O +i o o
t- CM « °
> > ^ 01
5SOO
j j 1 1
Table 4.351.1, continued
Displacement Due to Ocean Loading (cm in amplitude and degree in phase)
co
<
I
o.
0-
<
CO
<
I
a.
00 rv rv co cm ^ co to 01 t- i- m 05 m 03
-- i~ CM ^
|v Iv. |v |v tv
n o ■- m»
id cd h cd Ifl
co in m cm cd
en CO CD co co
rv rv iv rv rv rv
O0 00 ^f CO
cm *T 0"i in
co co
CO CO t- — CO
to n to in (o o> ca to o ^
d ^ d cm' to ^ d in d in
LO |v in t- CO (D CU O Tf N
t- 1- a) oi
c\i d ai 0 <d
co cm o r- in
in m o to to to o
CO c\i
to CO
I
co en co cn i-omo>o rv m
d d cm c\i d d d d ■<*■ r-' co
MU? r r N N CM CO t-
r> »- t- cm cm ,- ^ cm o 10 o n 01 jonnn 15153*0 g-co
00000 00000 00000 oO'-oo qoooo ppppp p p
CO
<
X
0.
EL
5
<£
CO
<
X
11
W \T CO O: W CM CO CO CO CM CO to in rv tv CO t-; rv (0 O W (fl t- CM CM q rv CO Ifl r
o o rv co
CO CO
I I
to CO
I
CO CO CM
CO CO CO
I I I
o ■*
CM CM CM ^ CM cm cm cm co to n n n n J OOlCMinCO O^CVOJIO COCOCOCOt- t- ^
OOpOp OppOO OOOOO t- d O r- T-pppp ppppp p p
cupT-coco cn co cm co cn co co cn *f rv *f to co to p q q q <p * 01 id q n
d rv' iv' d ^cdcoco'oi co tdiriiriiri to d d d cm ino'co^o S 21 12 SB ^ ^ £S
h,-V=t-t-CM t- t- CM OJ CM CM CM (M CM CO IV rv m O (D in I& o ^ co c\j co co in ^1-
" T T " 1 77 1 1
cd cd cm to m 01 n n co co co cn m co cn o cn o co 05 n r- 9 9 r 1: ? ^
1— 1- i- ,-0'-'-'^ — o — q o qqqqq *- q
s
I
CL
0.
<
<
I
CL
Cl
<
in
<
CL
0-
S
<
<
1
CL
CO
<
X
0.
<
CO
<
CL
co
<
X
a.
a.
<
*t CM lO CM CO CMTftOptD r- (v ^7. I1*. fv_ Is.
S--'*cOffi co'cM^ap'iri g ^ *r' *r' P co^-^-oot-
uffitoai loiouin'* n j | »- iff w jjfl
co |v tv co iv
O O CO r Cli
co to rv o co
(D s in CM
CM CM in o
»» *- CM CO
I I I I
in cm
o> CO
I I
cm cn — t- ^ cm in cm to co n id s cm r- jnooj £s!£!S32 2JS
o o *- o o 00000 cmocmt-cm n o cm o ooppp op
Q CM CO CO t- *- CO i- t- CO
r-^ Cli CM r4 CM CO
Tj- ■* tj- *■ m in ^
co to T— s to O) cn co
odopoS tfj iri to t- iri
■d" -«t CO CD O) !D r-
CMcomcOf- s co ^ to r- q
r> cd cd s 'j t^r^iriLrioS co 10
cm o co w m ^ m m o i-m
T *• T *" I I I I I *• 1
lo co co eg co ^- cm 3- in s 01 n r> co <j> cm co ^ o ui n cn fc £: S£ SS S S
^t^Tfcoco ^ 10 10 co to •? 9 t in « # u n 9 co - co o o q o q o -r- cm q
co m to ^ m n q ifi in
05 to cd r- d co Lri in c\i iri
iflinininco m1*^ 1
I I I I 1 III
r CO CJ> (S N
ic 0 fti c\i cd
CM 1- 1- j
Jill
^ m in co oj
W N N CD CO
in iri ^ iri cd
CH" CD CO CO
I T I
CM CO t— CM O
■t ^ ^ W N
CO CO CO CO J!
Mil
CO
I
in CO CO CO O CM CO CO CM ^ CO CO CM CM N H O r r CO CO 0) CO CO f 00 CQ ^ ^ O fell
CM CM CM -i- ^ CM CM CM — i- CM CM CM CM t If) CM CO if t- r O 1- 0 0^^01 O T_
Ol CO Ol »- CM
in in co co
CM OJ CM CM CO
N CM CO Tt N
CD N N N CO
CM CM CM CO CO
cn t- cm cm m
If) ^ Tf
CO CO CO CO CO
O ■»- h- CM f- CM CM CO CO S
CO CM LQ O T-
CO CO CM CO iO
O O CD O) iC
o
05
(0
CO
CD
CO
to
CO
X
t.03
1.04
CD
CD
cn
LO
orj
in
cn
LO
cn
1.05
CO
CO
uo
CO
1.05
to
to
1.06
a>
CO
o
00
CP
o
o
in
Is.
OJ
t
**
(
CD
CO
o
o>
cn
S
CO
4
u
$
CO
CM
m
CO
■*
cn
CO
'?
r-
00
CO
cn
a}
to
CM-
CM
r-
n
to
CO
in
CN
CM
o
CO
C5
as
CD
CO
Id
Ifl
LO
CO
LO
ro
tn
cn
C\J
n.
P
Ifl
CD
10
LO
CO
CM
CO
K-
CO
00
in
m
*t
CO
o
OS
gj
CD
1
CM
N
1
CO
1
s
1
in
in
7
ri
|v
1'
OJ
LO
1
CO
If]
1
o
cn
CO
cn
1
CO
CO
LO
o
s
CO
K
CO
i
r;
d
CO
7
d
CM
r-
n
to
CO
CD
s
ft
m
ID
o
cn
CM
CO
o
cd cn co
CO CM CO i- O
en to co <o cn
r- in *q-
N ^ ^ CO N
t-' s cci oi oi
CO CM CM *~ CO
I I I I
§ S IS i§ CO
CO CM tO CM CO
7 i i
(O O CO CO ^
d oj co co to
CM CO CO CO
G> tj- tj- tj- o> cn co
hw CO S O O i-; tJ; K
CO
CD
rs
P
t
>*
CO
CO
CO
in
r-
OJ
o.
OJ
CO
CM
3
CD
CD
7
1
7
i
1
1
OJ
LO
CO
OJ
a
cn-
CO
ro
q
s
iq
P
CO
CM
CO
0)
CM
o
CM
LO
in
CO
Bi
i
o
m
1
d
LO
1
cd
cn
1
CD
CO
s
1
CO
81
o
OJ
OJ
ID
CM
CO
CD
CD
CM
LO
CD
in
CO
T
CD
CD
ai
ro
CO
•I
■D
E5
CO
1
OJ
1
OJ
1
Oj
CO
o
7
i
o
o
0)
CO
CO
CO
S S CD Ol ^
f 5 =5 E I
> m to o
£ 2 co CD S
9= « I 8 u
o > a- 1
s *
_ ^ to o ca to
a I u. > cn
■S - °
.Sun.
i? O
nj = cj ro t_
? C3 iC >£ CD
IhiC
B H -
ill
crj co = _q ^
g g i | 1
O O S m O
2 3
23
TohiP 4 351 .1 , continued
□^placement Due to Ocean Loading (cm in amplitude and degree in phase)
it
s
cr?
L
n
D.
CO
<
X
S
^~
0-
2
^
<
X
o-
s
cn
<
I
°-
Q.
S
<c
CO
<
I
oT
CL
0.
CO
<
X
a.
CL
5
<
-p
o
CL
CL
<
CO
<
X
_
CL
<
CO
<
X
CL
CL
2
<
CO
<
I
Cl
Q.
<
0|s«^CO*<3" CO CM ^
r- cd' in ai fi ^ to 3- in
enh-r^enr- s co co oi
m co o m o
co o o) cs id Tf rj) in in < s <m. q in
_ £ *a _j in in co if) cji t— cm t- m
N- cm co t- o cn cn co
LO C7l t- iD
i— CO OJ co to
i- q q q q q
to i- to m in
o O o q q
w6HDNO coommto Noh-ioo scq to in « r^cor^cn^ ^ cm q cm r-
Lf) t- r- T" CO ]" r t\|C0 if)
qqqqq qqqqq
co co o co ^ co m cm o <o ffl r cm n co qcqcMr^cM qoi-rtqs S^^^^
■ ^ ■ _- .■ _ ™< m n m ^ » rrt ^ N i t— h- 1^ GO CO CO Ifl O CD r-
r^cdcdcMto oi^tcooico co
O CM CO
*- co m
1 1 7
* in
p °.
ffiCM*-OJtp '-T-cocom
ppppp ppppp
. oinfl9 n in 4 to cd cocnocDO r-, cm p cn r^pppcM pptgin*
fi i | 7 llll77 1 7T 1 i
55585553555 S^^SSSSpo. q&?8.;5?555
CO CO CO »- f** "vf CO CD
^ t>j CO CM r-' J *
CO CO CM CO 't N
I I I I I I
^th^ CO CO cn O CD OJ 't N o o
CM CO
I [
S CO aj if) T- IN oj oi to o
5 O co i r- Cn ^- CO
fl) CO if) ^ t-
cm cd cm in
CO O CO S CO
CO O CO -J i-
co *— o 0"i c\i
m v 1- i
§ ^ S 1$ S li 8. 5i J? I| S S 5 5 Q 5 % S,
cMT-ppp 3P^pp rAj^cqin.cq pr^ppp to cqp^p r^^pptn
-64.2
-54.1
-53.6
174.8
-54.0
-48.4
-58.3
si
CO
-52.3
cn
to
-55.2
LO
CO
126.6
LO
iO
C'J
OJ
CO
I
oi
cn
oj
in
1
d d
cn in
1 1
to
1
o
CO
OJ
lO
cd cd
CO CO
LO
fv*
|v
LO
CO
o
LO
O
LO
o
I
cm m
P °.
L0
O
LO
q
p
P
1 I
q q 3
LO
OJ
CO
o
CO
m
lO
p
r-
p
to
q
go
o
L0
CO
o-
to
Ifl
CO
m
CO
CM
o
s
CO
o
OJ
CM
d to
no co
I I
CM
CO
in
CM
CO
Clj
CO
r*.
LO
ro
to
CO
CO
I
LO
m
s
CM
in
00
1
T
od
oj
7
>Jr
CO
d
CI
Ifl
LO
1
q
K
C\J
|v
to
oi
1
O
rv
CD
IV
a
Cn
I
cn"
CO
I
S5
I
en
CO
1
CO
00
1
CM
OJ
s
1
5
1
8
i
5
CO
CO
CD
s
1
^0
1
i-
1
i
s
1
7
CO
1
03
1
a)
cn t-
U0
Ifl
7
a
o
CM
LO
in
o-
CO
cn
tn
CO
cn
CJ
s
CO
CM
CO
CM
O
C-J
<C
o
8
CO
00
Oi
I
cn
m m
cn
OJ
o
cn
00
CO
CO
ro
1
Ifl
OJ
LO
cn
CO
1
CM
CO
OJ
ro
LO
cn
ED
q
CM
r-.
CO
OJ
cn
CM
ro
m
OJ
LO
CM
tv
Tf
OJ
OJ
LO
o
CO
Cd
d
m cn
o-
cn
1
CM
CO
1
CO
69
1
1
Si
Ifl
LO
d
to
ro
a
1
ro
1
7
7
a
CO
o
7
7
CO
o
CO
o
co
p
7 '
in co
p o
7
CD
q
o
s
7
in
p
q
1
CD
I
CO
CO
CD
in
o
P
1
to
CM
s
q
s
o
S!
CO
OJ
q
to
Si
co
IV
q
cn
CO
£
Ol
o
-71.8
CO
to
I
CO
Is
In
I
cn co
d K
is. m
CM
0J
CO
|
o
CM
CD
1
cn
cm
ID
L0
in
1
in
Lfj
1
O
ID
|
CM
in
CO
to
CD
CM
7
o
d
OJ
CO
CO
1
an
N.
ro
1
o
IT
p
m
i
m
ro
qd
1
CM
ai
in
1
CM
d
CD
T
Ifl
oj
to
to
4
co
OJ
1
IN
|v
LO
ecj
SB
CO
in
cn
CO
to
to
CO
d
cc
1
■J
to
1
a
r-
s
7 1
CO CO
p I-
N
o
cn
O
m
1
ro
5
q
cn
OJ
CO
7
o
LO
s
«
o
CO
?
ro
q
CO
CO
CM
CM
■*
|v
p
CM
-39.9
m
co
co to
CO
r-
CO
O
o
OJ
CM
CM
CO
CD
CM
0-
CO
ro
cn
IO
q
CO
1
CO
CD
7
CC
d
to
|v
CC'
f-'
to
I
CO
CO
I
Cli cd
in co
1 1
CO
CM
1
co
1
CD
CO
1
3
i
cn
CO
1
r-
o
CO
CO
I
rv"
1
CO
7
LO
cn
S)
CM
1
5
I
1
o-
1
CM
7
o
7
CO
7
7
L0
1
7
**
— V
CO
to
CO
1-
CO
CO
<«
a
CM
15
LO
ia
m
9
CO
OJ
s
-63.9
LO
CD
I
i
O CO
co
ID co
1 I
00
ro
LO
if
1
m
o
CO
CO
to
1
CO
S
Ol
s
1
LO
9
CD
O-
CO
CM
a;
CD
CO
CO
CO1
1
cn
Ol
CD
CM
^J
CM
CO
1
is
LO
o
7
cn
s
1
o
cc
CO
CO
7
d
7
cn
lO
7
i
■3
CO
CO
in
tri
m
1
CO
>n
1
CM
CM
in
o
in
CO
co in
CO
CM
CO
0-
7
CO
V
iO
cn
■a-
CM
r-
in
0J
O-
o
CO
CO
LO
!a
CO
3
cn
CM
CM
co
q
3
CO
c
Sj
i
m
CO
LO
CO
C\
'OJ
CO
as
CO
10 r- _ E
c — -o S d i2ScoiSc 2 tj r n o coacicp ir, oSto o«
Ei6i| IcSlCLO ZOiilSi co<cfitON ILtSOS 20.
248 EXPLANATORY SUPPLEMENT
The height displacement, in centimeters, can be found for a given constituent /' as
follows:
h(i) = amp(r) x cos(arg((", t) - phase(i)), (4.351-2)
where arg(i,f) is the proper angular argument to be used with the Schwiderski
phases (a Fortran subroutine for this exists in the IERS Standards (McCarthy,
1989, pp. 40-42)). The total tidal displacement is found by summing over all nine
constituents. A negative h(i) means that the surface at the predicted point has been
lowered.
4.352 Atmospheric Loading The procedure described below is taken from the pub-
lication of Sovers and Fanselow (1987). A time- varying atmospheric pressure dis-
tribution can induce crustal deformation. Rabbel and Schuh (1986) estimate the
effects of atmospheric loading on VLBI baseline determinations, and conclude that
they may amount to many millimeters of seasonal variation. In contrast, to ocean-
tidal effects, analysis of the situation in the atmospheric case does not benefit from
the presence of a well-understood periodic driving force. Otherwise, estimation of
atmospheric loading via Green's function techniques is analogous to methods used
to calculate ocean-loading effects. Rabbel and Schuh recommend a simplified form
of the dependence of the vertical crustal displacement on pressure distribution. It
involves only the instantaneous pressure at the site in question, and an average
pressure over a circular region C with a 2000-km radius surrounding the site. The
expression for the vertical displacement (mm) is
Ar = -0.35p~0. 55p. (4.352- 1 )
where p is the local pressure anomaly, and p the pressure anomaly within the 2000-
km circular region mentioned above (both quantities are in mbar). Note that the
reference point for this displacement is the site location at standard pressure (1013
mbar) .
An additional mechanism for characterizing p may be applied. The two-dimensional
surface pressure distribution surrounding a site is described by
p(x,y) = Aq + A\x + A2y + A^x2 + A^xy + Asy2 , (4.352-2)
where x and y are the local east and north distances of the point in question from
the VLBI site. The pressure anomaly p may be evaluated by the simple integration
= j I dxdyp(x,y) / J fdxdfc (4.352-3)
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 249
giving
p=A0 + (A)+A<,)R1 14, (4.352-4)
ffhere/?2 = (^+/)-
It remains the task of the data analyst to perform a quadratic fit to the avail-
able weather data to determine the coefficients A0_s- Future advances in under-
standing the atmosphere-crust elastic interaction can probably be accommodated
by adjusting the coefficients in Equation 4.352 1. Furthermore, expansion of Equa-
tion 4.352-1 might be required for stations close to the coast.
4,36 Plate Motions
One of the factors that can affect Earth rotation results is the mot ion of the tectonic
plates that make up the Earth's surface. As the plates move, fixed coordinates
for the observing stations will become inconsistent with each other. The rates of
relative motions for some regular observing sites are believed to be 5 centimeters
per year or larger. The observations of plate motions so far by satellite laser-ranging
and Very Long Baseline Interferometry appear to be roughly consistent with the
average rates over the last few million years derived from the geological record
and other geophysical information. Thus, in order to reduce inconsistencies in the
station coordinates and to make the results from different techniques more directly
comparable, a model for plate motions based on the relative plate motion model
RM-2 of Minster and Jordan (1978) is recommended.
From the RM-2 model, Minster and Jordan derive four different absolute plate
motion models. The two that have been discussed most widely are AMO-2, which has
zero net rotation of the Earth's surface, and AMl-2, which minimizes the motion of
a set of hot spots. Kaula (1975) and others have discussed alternative geophysical
constraints that can be used in order to investigate the plate motions with respect
to the bulk of the mantle. For convenience, when future changes are made in the
relative-motion model, and to avoid dependence on the choice of the hot spots to
be held fixed, the (AMO-2) model is recommended. The cartesian rotation vector
for each of the major plates is given in Table 4.36.1.
Future improvements are planned for future IERS Standards including: (1)
replacement of RM-2 by a new model, new velocities, and new plates— for example
with the NUVEL-1 model of DeMets et al. (1990); and (2) adoption of a vertical-
motion model (post-glacial rebound). In any case, the AMO-2 model should be
used as a default, for stations that appear to follow reasonably its values. For
some stations, particularly in the vicinity of plate boundaries, users may benefit
by estimating velocities or using specific values not derived from AMO-2. This is
also a way to take into account now some nonnegligible vertical motions. Published
station coordinates should include the epoch associated with the coordinates.
250
EXPLANATORY SUPPLEMENT
Table 4.36.1
Cartesian Rotation Vector for Each Plate Using
Kinematic Plate Model AM0-2 (No Net Rotation)
nx
Plate Name
deg/My
deg/My
deg/My
Pacific
-0.12276
0.31163
-0.65537
Cocos
-0.63726
-1.33142
0.72556
Nazca
-0.09086
-0.53281
0.63061
Caribbean
-0.02787
-0.05661
0.10780
South America
-0.05604
-0.10672
-0.08642
Antarctica
-0.05286
-0.09492
0.21570
India/Australia
0.48372
0.2501 1
0.43132
Africa
0.05660
-0.19249
0.24016
Arabia
0.27885
-0.16744
0.37359
Eurasia
-0.03071
-0.15865
0.19605
North America
0.03299
-0.22828
-0.01427
4.37 Tidal Effects on UT1
To properly compare UT1 results obtained at different epochs, the high-frequency
tidal effects must be accounted for. Thus, a standard expression for the tidal vari-
ations in UT1 is required. The periodic variations in UTl due to tidal deformation
of the polar moment of inertia have been rederived (Yoder et ai, 1981) including
the tidal deformation of the Earth with a decoupled core.
This model leads to effective Love numbers that differ from the bulk value of
0.301 because of the oceans and the fluid core, which give rise to different theoretical
values of the ratio kl C for the fortnightly and monthly terms. However, Yoder et ai,
recommend the value of 0.94 for k/C for both cases. Periodic terms in UTl are given
in Table 4.37.1 and are discussed in Explanatory Supplement to the IERS Bulletins
A and B, March 1989 as well as the Annual Reports of the IERS. Table 4.37.1
includes terms due to zonal tides, with periods up to 35 days, with k/C = 0.94.
Table 4.37.2 lists the remainder of the tidal terms (with periods greater than 35
days).
UTlR, AR, and ujR represent the regularized forms of UTl (including periods
up to 35 days), of the duration of the day, A, and of the angular velocity of the
Earth, w. The units are 10-4s for UT, 10_5s for A, and 10_I4rad/s for 0.
It should be mentioned that the oceanic tides with short periods cause varia-
tions in UTl that are only partially represented by the model given below. Accord-
ing to Brosche et al. (1989) the contribution of the oceanic tides should be split into
a part that is in phase with the solid Earth tides and into an out-of-phase part. The
discrepancy between the model of Yoder et ai, (which already contains a general
influence of the oceans) and the model of Brosche et al, amounts to a maximum
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 251
0f ± 0. 08 msec for the fortnightly and monthly terms. The influence of the ocean
tides with periods less than or equal to 24h is at the level of ± 0. 1 msec.
4 4 THE MONITORING OF THE ROTATION OF THE EARTH
Continuous monitoring of the Earth's rotation began after Kiistner's discovery of
the variation of latitude in 1884, and Chandler's explanation of it as polar mo-
tion (Chandler wobble). Optical astrometric methods then in use have continued
to this day, and in fact continued to be the only source of Earth-rotation infor-
mation up until the mid-1960s. More modern techniques then gradually took over
the bulk of the activity of monitoring the rotation of the Earth, with successively
Doppler tracking of satellites, Connected Element Radio Interferometry (CERI),
Lunar Laser-Ranging (LLR), Satellite Laser-Ranging (SLR), and Very Long Base-
line Interferometry (VLBI) assisting in such monitoring. At present, the techniques
of SLR and VLBI provide most of the current high-accuracy observations for this
purpose, with LLR providing such data sporadically.
Before we describe the observational systems and services, the basic sensitivity
of the systems should be briefly explained. All the current observational systems
depend on the existence of observing astronomical targets whose position can be
inferred in some celestial reference frame. This includes stars and solar-system ob-
jects for optical astrometry, the Moon and artificial satellites for laser ranging, and
(usually distant) radio sources for radio interferometry. The observations of these
systems are then used to infer the motion of the observing instruments or, more
specifically, the Earth to which they are attached. After the predicted motion from
all other "defined" sources (as explained previously, including precession, nutation,
the Earth's sidereal rotation, tides, tectonic motions, etc.) has been removed, the
motion that remains is attributed to polar motion and UTl-UTC variations. How-
ever, as we shall see, some proposed fut ure systems of Earth rotation monitoring do
depend instead on the sensing of a local "inertial" coordinate system, rather than
an astronomically defined one.
It is also important to note that a single instrument (or single baseline for in-
terferometric observing), being at a given point on the Earth's surface, can be used
to determine only two components of the Earth's rotation. The components com-
monly determined using single-station (or single-baseline) data include "variation
in latitude" (A<j>) due to any north-south motion of the station, and UT0-UTC
(or just UT0) when there is any east-west motion of the station, attributable to
Earth rotation variations. To obtain all three components of Earth rotation, it is
necessary and sufficient to use data from two stations (or baselines) with different
longitudes. If multiple stations (or baselines) with different longitudes are involved,
the problem is overdetermined and generally a least-squares solution is done to
determine the Earth-rotation parameters (ERP).
Table 4.37.1
Zona! Tide Terms with Periods Up to 35 Days
Argument* Period
/ I' F D a Days
UT1 - UT1 R A-AR u> - u;R
Coefficient of Coefficient of
sin (argument) cos (argument)
1
0
2
2
2
5.64
— 0.02
Lr.o
— U.^
2
0
2
0
1
6.85
-0.04
0.4
-0.3
2
0
2
0
2
6.86
-0.10
0.9
-0.8
0
0
2
2
¥ '
7.09
-0.05
0.4
-0.4
0
0
2
2
2
7.10
-0.12
1.1
-0.9
1
0
2
0
0
9.11
— 0.04
n o
U.d
—0.2
1
0
2
0
&
9.12
-0.41
2.8
-2.4
1
0
2
0
2
9.13
-0.99
6.8
-5.8
3
0
0
0
0
9.18
-0.02
0.1
-0.1
— 1
o
2
2
1
9.54
-0.08
0.5
-0.5
— 1
0
2
2
2
9.56
—0.20
1 .3
— 1.1
1
0
0
2
0
9.61
-0.08
0.5
-0.4
2
0
2
-2
2
12.81
0.02
-0.1
0.1
0
1
2
0
2
13,17
0.03
-0.1
0.1
0
0
2
0
0
13.61
-0.30
1.4
-1.2
0
0
2
0
1
13.63
3.21
14,8
— 12.5
o
o
2
0
2
13.66
-7.76
35.7
-30.1
2
0
0
0
1
13.75
0.02
-0,1
0.1
2
0
0
0
0
13.78
-0.34
1.5
-1.3
2
0
0
0
1
13.81
0.02
-0.1
0.1
0
—1
2
0
2
14.19
—0.02
0,1
—0.1
0
0
0
2
-1
14.73
0.05
-0.2
0.2
0
0
0
2
0
14.77
-0.73
3.1
-2.6
0
0
0
2
1
14.80
-0.05
0.2
-0.2
0
-1
0
2
0
15.39
-0.05
0.2
-0.2
\
0
2
-2
1
23.86
0.05
—0.1
0.1
0
2
-2
2
23.94
0.10
-0.3
0.2
1
1
0
0
0
25.62
004
-0.1
0.1
_■)
0
2
0
0
26.88
0.05
-0.1
0.1
_1
0
2
0
1
26.98
0.18
-0.4
0.3
-1
0
2
0
2
27.09
0.44
-1.0
0.9
0
0
0
-1
27.44
0.53
-1.2
1.0
0
0
0
0
27.56
-8.26
18.8
-15.9
0
0
0
1
27.67
0.54
-1.2
1.0
0
0
1
0
29.53
0.05
-0.1
0.1
-1
0
0
0
29.80
-0.06
0.1
-0.1
0
0
2
-1
31.66
0.12
-0.2
0.2
0
0
3
0
31.81
-1.82
3.6
-3.0
0
0
2
1
31.96
0.13
-0.3
0.2
0
-2
2
-1
32.61
0.02
0.0
0.0
-1
0
2
0
34.85
-0.09
0.2
-0.1
* / = 134. 96° + 13. 064993°(MJO-51544.5) Mean anomaly of 1he Moon
/' = 357.53° + 0. 985600° (MJD-51544.5) Mean anomaly of the Sun
F = 93.27° + 13. 229350° (MJD-51544.5) L- SI:U Mean longitude of the Moon
D = 297. 85° + 12. 190749° (MJD-51544.5) Mean elongation of the Moon from the Sun
SI = 125. 04° - 0.052954° (MJD-51544.5) Mean longitude of the ascending node of the Moon
/ TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 253
Table 4.37.2
Zonal Tide Terms with Periods Greater than 35 Days
Argument*
Period
UT1 - UTIfl
A-AR
u- - ujR
<—
— —
Coefficient of
Coefficient of
/
/'
F
D
n
Days
sin (argument)
cos (argument)
0
2
2
-2
2
91.31
-0.06
0.0
0.0
0
1
2
-2
1
119.61
0.03
0.0
0.0
0
1
2
-2
2
121.75
— 1.88
1.0
-0.8
0
u
2
_2
o
173.31
0.25
-0.1
0.1
0
f\
u
2
-2
1
177.84
1.17
-0.4
0.3
0
0
2
-2
2
182,62
-48.25
16,6
-14.0
0
2
0
0
0
182.63
-0.19
0.1
-0.1
2
0
0
-2
-1
199,84
0.05
0.0
0.0
2
u
o
_2
o
205.89
-0.55
0.2
-0.1
2
u
o
-2
1
212.32
0.04
0.0
0.0
0
_1
2
-2
1
346.60
-0.05
n rv
U.Lf
0
1
0
0
-1
346.64
0.09
0.0
0.0
0
-1
2
-2
2
365.22
0.83
-0.1
0.1
0
0
0
0
365.26
-15.36
2.6
-2.2
0
0
0
1
386.00
—0,14
0.0
0.0
1
0
0
-1
0
411.78
0.03
0.0
0.0
2
0
-2
0
0
1095,17
-0.14
0.0
0.0
-2
0
2
0
1
1305.47
0.42
0.0
0.0
-1
1
0
1
0
3232.85
0.04
0.0
0.0
0
0
0
0
2
3399.18
7.90
0.1
-0.1
0
0
0
0
1
6790.36
-1617.27
-14.9
12.6
/ = 134. 96° + 13. 064993°{MJD-51544.5) Mean anomaly of the Moon
/' = 357. 53° + 0. 985600° (MJD-51544.5) Mean anomaly of the Sun
F = 93.27° + 13. 229350° (MJD-51544.5) L - Si: L; Mean longitude of the Moon
D = 297. 85° + 12. 190749°(MJD-51544.5) Mean elongation of the Moon from the Sun
n = 125.04° - 0.052954° (MJD-51544.5) Mean longitude of the ascending node of the Moon
The actual equations for expressing Acp and UT0-UTC in terms of xp, yp, and
UT1-UTC are
A<j> = xp cos A - yp sin A, (4.4-1 )
UT0 - UTC = tan <Mx„ sin A + yp cos A) + (UT1 - UTC), (4.4-2)
where r/> and A are the latitude and longitude of the station (or baseline), with
longitude again measured toward the east. In cases where data from more than
two stations (baselines) exist, a least-squares solution is generally performed with
the above equations (for most multistation optical astrometry and LLR quick-look
solutions), or else a general least-squares solution is performed, solving for the three
components of Earth rotation (among many other parameters) directly from the
observations.
254 EXPLANATORY SUPPLEMENT
Following are discussions of the individual methods for Earth-rotation deter-
mination. First, the current high-accuracy methods are described including laser
ranging (both lunar and satellite) and VLBL Second, historical methods are dis-
cussed. Third, some proposed future or alternate techniques are described. For all
these methods, only basic descriptions of the observing systems are presented, along
with summaries of the actual operational systems. Finally, the current international
service for Earth rotation is described. (A similar review also exists in Moritz and
Mueller, 1987, as well as in IERS, 1988.)
4.41 Laser Ranging
Laser ranging of celestial targets, first proposed (for the Moon) in the late 1950s and
early 1960s (Alley, 1983), is conceptually a very simple observational procedure. A
laser beam is directed to a celestial target carrying retroreflectors, such as the Moon
or an artificial satellite. Half the round-trip travel time of the light multiplied by the
velocity of light gives the distance to the object (subject to various corrections due
to relative motion, refraction, etc.). These ranges can then be compared to their
predicted values, and in an adjustment (usually weighted least squares), corrections
to the parameters involved may be estimated, such as the orbital parameters, the
station positions, and the orientation of the Earth.
4.411 Satellite Laser-Ranging Currently there are several satellites available for
laser ranging, but observations of the Laser Geodynamics Satellite (LAGEOS) cur-
rently provide the highest-accuracy values for the above parameters. (However, the
USSR has recently launched two similar satellites, Etalon 1 and 2, which show great
promise as targets for providing such information (Anodina and Prilepin, 1989).)
LAGEOS itself is a composite (aluminum shell and brass core) sphere, 60 cm in
diameter, weighing 407 kg, with 4263.8 cm cube corner retroreflectors embedded in
its surface. It is in a nearly circular orbit at a height of about 5900 km, and an
inclination of 109.8°. Its relatively large mass, small surface area, and high orbit
result in low atmospheric drag and insensitivity to short-wavelength components of
the Earth's gravity field. More information on LAGEOS and other such geodetic
satellites is available in Cohen et al. (1985).
There are approximately 25 SLR stations in regular operation, located on eight
tectonic plates, regularly obtaining observational data. The precision of ranges ob-
tained from these instruments are on the order of ± 1 to ± 15 cm. (For more infor-
mation on the instruments themselves, see Shawe and Adelman, 1985 and Degnen,
1985.) These stations are operated by various governmental and university research
organizations, in order to obtain various geophysical data relating not only to Earth
rotation, but station coordinates and baseline lengths, changes in those values over
time {thereby giving tectonic-plate motion velocities), and long- wavelength compo-
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 255
nents of the Earth's gravity field. The coordinates of the stations computed by this
technique have an origin which is thought to be coincident with the center of mass
of the Earth to within! 10 cm. Earth-rotation results are computed at 3-day inter-
vals using data less than a week old, with polar motion and UT1-UTC accurate
to a few milliseconds of arc (mas) and a few tenths of a millisecond respectively,
(Further information on the computational techniques involved is given in Kaula,
1966, pp- 78-81, and GSFC, 1976, pp. 7-11 to 7-15.)
4.412 Lunar Laser-Ranging Lunar Laser-Ranging (LLR) is identical in concept
with SLR, except that the targets are arrays of cube corner reflectors left on the
Moon by the U.S. Apollo astronauts and USSR unmanned probes. However, the
distance to the Moon makes it more difficult to point a laser reliably at the re-
flectors, and when there are return signals, they are orders of magnitudes weaker
than those from the LAGEOS satellite. Although LLR has been done since Au-
gust of 1969 (initially using the array left by the Apollo 11 astronauts (Armstrong,
1977)), these technical difficulties have resulted in few stations reaching continued
operational status. Currently data is obtained regularly by the Centre d'Etudes et
des Recherches Geodynamics et Astronomiques (CERGA) station near Toulouse,
France; the University of Hawaii LURE Observatory on Maui, Hawaii; and the Mc-
Donald Observatory, Texas. Much of the same information obtained with SLR can
be obtained from LLR. However, in practice, with only a limited number of stations,
the primary use of the LLR systems has been for the determination of UT0-UTC
(and thereby UTl-UTC), parameters of the Moon's orbit and the Moon's rotation
(librations), special- and general-relativity parameter tests, and information on the
tidal parameters of the Earth-Moon system. LLR also shows promise in providing
quick determinations of UT0-UTC, since its computation is a simple matter as soon
as the day's observations are completed at a given station. Currently UT0-UTC is
determined irregularly (at each of the three operating stations) with an accuracy
of ± 0. 3 to ± 0. 7 ms (McCarthy and Luzum, 1990). Larden (1982, p. 250) indicates
that with a four or five station network, and ranging regularly with ± 3 cm range
accuracy, that ± 5 cm (± 1 . 6 mas for polar motion, iO.lms for UTl-UTC) two-day
ERP determination should be possible. Further information on the computational
techniques involved is given by Arnold (1974, pp. 249-255), Mulholland (1977), and
Larden (1982, pp. 69-90).
4.42 Very Long Baseline Interferometry
In geodetic Very Long Baseline Interferometry (VLBI), the primary targets being
observed are a set of extragalactic radio sources. These objects are presumably so
distant that none of their motions are detectable, and they therefore form a fixed
celestial reference frame. Currently, some 228 sources have positions established
(published) within such a system, but only about 74 objects have been observed
256 EXPLANATORY SUPPLEMENT
from more than one network, and only about 23 repeatedly from more than two
networks (Arias, et ai, 1988).
The VLBI observing networks consist of two to generally four or five radio tele-
scopes that simultaneously track extragalactic radio sources. Each station records
X and S band radio signals in digital form on magnetic tape, along with precise
time signals provided by hydrogen-maser frequency standards. Typically, 10 to 20
sources are tracked for periods of 3 to 6 minutes several times over the course of a
24-hour observing session. The tapes are sent to a correlator center, where they are
replayed. The differences in arrival times of the radio source signals ("delays"), and
their differences in time ("delay rates") are determined between each pair of sta-
tions. Similar to the laser ranging solutions, these delays and delay rates are then
used in a least-squares adjustment in order to estimate a variety of geophysical
parameters and the coordinates of the radio sources.
Currently some 20 radio telescopes in the world have been outfitted for making
high-accuracy geodetic VLBI observations (using Mark III equipment, see Clark et
ai, 1985). Although many of these are radio telescopes shared between geodetic
and astrophysical observations, several are dedicated primarily to determination of
ERP. These include primarily two networks: Projects Polaris/IRIS (Carter et ai,
1984; Carter and Robertson, 1984), operated by the U.S. National Geodetic Survey,
consisting of four or five stations operating in the United States and Europe; and
the Navy Network, operated by the U.S. Naval Observatory, consisting of four to
six stations operating in North America and Hawaii. The IRIS stations observe
for 24 hours, every 7 days, using 14 different radio sources. Current accuracies
of the data obtained include polar motion at ± 1-2 mas, UTl— UTC at ± 0.05-
0.10 ms, nutation and precession (celestial pole) corrections at ± 1-2 mas, radio-
source coordinates at less than ± 1 mas, and VLBI station coordinates at a few
centimeters. Daily observations of four sources for 1 hour, from two stations, also
provides daily determinations of UT0-UTC. The Navy Network observes for 24
hours every 7 days, using 26 different radio sources. Observations every few days
for three hours provide daily UT0— UTC values (with two stations) with accuracies
similar to the IRIS results. Observations are also made by stations operated at
occasional intervals by the U.S. NASA Crustal Dynamics Project (CDP) for the
monitoring of plate tectonics and Earth rotation (Coates et ai, 1985).
Further information on the computational techniques involved is given by GSFC
(1976, pp. 7-41 to 7-42) Robertson (1975), Ma (1978), McLintock (1980), and Bock
(1980).
4.43 Historical Methods
Various methods of lesser accuracy for ERP determination have existed in the past,
with some still in operational use today for special purposes. These techniques in-
clude primarily optical astrometric methods, with space techniques such as Doppler
TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 257
ellite tracking and Connected-Element Radio Interferometry (CERI) having been
>d for operational purposes in the 1960s and 1970s.
4 431 Optical Astrometric Techniques An explanation of the motion of the Earth's
pole was first made by Chandler in the early 1890s {Chandler, 1891, 1892), using
transit circle observations by Kiistner dating from the late 1880s. Chandler, and
later others, also used various astrometric observations from transit circles, quad-
rants, and other visual optical instruments dating back to the early 1700s. The
1890s saw the establishment of the International Latitude Service (ILS), which op-
erated several stations at approximately 39° latitude until the 1970s. The use of
visual zenith tubes and the invention of photographic zenith tubes (in 1913 by
Ross) brought the measurement of polar motion to accuracy levels pretty much
maintained into the 1960s. The use of precise pendulum clocks in the 1930s allowed
the detection of seasonal variations in the length of day. Special Moon cameras were
also used from 1952 until the early 1960s in order to measure changes in the length
of day against ephemeris time derived from the observed position of the Moon. The
same equipment was also used to compare ephemeris time with atomic time and
to establish the definition of the second (Markowitz et ai, 1958; Markowitz, 1960).
The International Polar Motion Service (IPMS), following on the ILS, coordinated
and reduced the polar-motion observations. The Bureau International de THeure
(BIH) did the same for UT0-UTC related observations, and also eventually be-
gan to handle polar-motion data in order to provide rapid service UT1-UTC. The
optical techniques continued as the only source of ERP measurements until the
mid-1960s, when Doppler satellite tracking began to provide information on polar
motion. More than 100 optical instruments all over the world were at one time
in use for such measurements. Currently, a small number of such instruments are
still in operation, primarily as a backup to and as a possible check on the space
geodetic techniques, and to continue their useful long record of continuous observa-
tions. The primary disadvantages (for improved ERP determination) of the optical
astrometric techniques are their dependence on the local vertical as their primary
reference and the accuracy of the star catalogs they use to define a celestial refer-
ence system. The geodetic space-based techniques that have mostly replaced them
are little affected by variations in the local vertical and provide instead for a crust-
fixed three-dimensional reference system. Information on these instruments and
the various international services involved is given in reviews (Moritz and Mueller,
1987, pp. 324-337, 424-430, 456-465; Mueller, 1969, pp. 80-84; Lambeck, 1980, pp.
62-66; Munk and MacDonald, 1960, pp. 58-61; Bomford, 1971, pp. 312, 335-338.)
4.432 Doppler Satellite Tracking During the 1960s, Doppler tracking of the U.S.
Navy Navigational Satellite System (NNSS) began to provide polar-motion infor-
mation on the same level of accuracy or better than the optical techniques. The
258
EXPLANATORY SUPPLEMENT
ERP values obtained via this system were used by the BIH in combination with the
optically obtained values. However, this technique did suffer from the dependence
of the ERP values on the particular satellites in use and from long-term drifts,
and as SLR became available for polar motion monitoring in the late 1970s, began
to be phased out for active ERP use. An excellent review of the NNSS Doppler
satellite-tracking technique is given by Kouba (1983), and a review of ERP-related
use of such tracking is given by Moritz and Mueller (1987, pp. 338-344, 467-470).
4.433 Connected-Element Radio Interferometry One other method that enjoyed
some use in the late 1970s and 1980s was the use of Connected-Element Radio
Interferometry (CERI). CERI operates in principle in the same fashion as VLB1
(and, in fact, provided the initial tests of the VLBI principle), except that the radio
telescopes are physically connected by a wideband communications link, so that
the observations {delays and delay rates — see Section 4.42) are made in real time.
A least-squares solution to obtain ERP, station positions, and source positions can
then be made as soon as observing ends for the day. Although the speed at which
results could be made available was a strong point of CERI, the small size of the
arrays usually only allowed two ERP components to be determined simultaneously,
such as A4> and UTO-UTC. The U.S. Naval Observatory operated such an inter-
ferometer at Green Bank, West Virginia, from October 1978 until the end of 1987.
Polar motion and UTO-UTC could be determined to an accuracy of about 50 mas
and 4.6ms respectively (McCarthy et ai, 1985; Matsakis et at, 1986). Another
CERI was also operated for a short time (August-September, 1980) at Cambridge.
U.K. (Klepczynski, 1982). The eventual improvement in the turnaround time and
much higher accuracy of VLBI results led to the eventual demise of this technique
for ERP determination.
4.44 Alternative Techniques
Several observational methods are under development that show promise in the
field of Earth-rotation measurement. The current level of development of these
techniques of course varies greatly, and the future levels of deployment and accu-
racies achievable by these techniques is still not clear. The primary contender for
providing high-accuracy Earth-rotation information is the Global Positioning Sys-
tem, a global network of 24 satellites intended for navigation, which is currently
only partially deployed. The use of laser ring gyroscopes or superfluid helium inter-
ference gyroscopes may also allow such changes in the Earth's rotational velocity
to be monitored. These last techniques may be of particular interest, since they
have no dependence on astronomical observations and may also be used as a check
on current techniques. What follows are more detailed accounts of these three pos-
4 i TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 259
sible alternative observational systems. Of course, it should also be realized that
improvements in the currently used observational systems (previously mentioned)
also show great promise.
4 441 The Global Positioning System (GPS) The Global Positioning System was
designed as an all-weather, continuous, and global satellite navigation system. In
its basic configuration it was to be used for navigation, but is also quite usable for
time transfer and geodetic positioning of points on the ground. It was designed, is
under development by, and will be operated by the U.S. Department of Defense.
The "space segment" of the system will consist of 24 satellites in 12-hour,
20000-km-altitude orbits, with the satellites equally spaced in three orbital planes,
inclined at 55° to the equator. Each satellite will broadcast on two L-band frequen-
cies, L\ (1575.4 MHz) and Li (1227.6 MHz), with the two frequencies allowing the
elimination of first-order ionospheric refraction effects. Modulated onto these fre-
quencies will be three codes, including (1) a "D" code, containing information (at
50 bits/second) on the predicted position of the satellite and the satellite itself, (2)
a coarse acquisition "C/A" code, a pseudorandom noise (PRN) code at a frequency
of 1.022 MHz, which repeats every one millisecond, and (3) a precise "P" code,
another PRN code at a frequency of 10.23 MHz, which repeats every 38 weeks, but
will be reset in practice every week. Each satellite also has a very precise cesium or
rubidium clock to provide precise time and stable frequency information. Currently
(1991 September), there are six operating prototype (Block I) satellites and eleven
operational (Block II) satellites in orbit that allow the system to be used inter-
mittently. The launch of more Block II GPS satellites is planned to continue using
Delta-2 boosters. This process will continue until the system is fully operational
(with 21 satellites and 3 spares) in 1993.
The "user segment" will generally consist of users with navigation receivers.
These receivers will be capable of "locking on" to four satellites at a time, gener-
ating C/A- (or for more precision, P-) codes, shifting those codes in time to match
those of the satellites, and decoding the D-code information from the satellites.
The position of the satellites will then be known from the D-code information, and
it is only necessary to take the four observations (one to each satellite) and solve
four equations for the four unknowns of the receiver's position (say X, Y, Z) and
time (this process being known as "pseudoranging" ) . A "Standard Positioning Ser-
vice" will be available to all users. This service allows use of the C/A code (but
with a degraded signal for the Block II satellites —in a process known as "Selective
Availability" (SA)), for determining the receiver's position anywhere on Earth to
approximately 100 200 meters. Using the P code, which will be available only to
qualified users as the "Precise Positioning Service," positional accuracies of 15-30
meters are possible. Once "locked on" to the satellites (a process that takes at most
260 EXPLANATORY SUPPLEMENT
a few minutes), real-time updates will be available every 4-5 seconds. Differential
techniques are under development by the private sector that will eventually al-
low instantaneous cm-level relative positioning. In these techniques it is envisioned
that permanent ground stations will continuously track GPS satellites and supply
their observations (in some form) over a communications link to mobile users. The
observations may then be "differenced" or combined in some way to get very high-
accuracy station positions. Since any change in the satellite signals will affect both
receivers equally, any degradation or coding of the signals will have little effect on
the accuracy of the relative station positions.
For geodetic use of these satellites, observations can simply be made for longer
periods, and with more satellites, in order to increase the positional accuracy. It is
estimated that absolute accuracies on the order of + 1 meter or less are possible.
Alternatively, two or more receivers may also be used in a differential or interfer-
ometric mode, analogous to the VLBI observations discussed previously, and the
relative positions of the two receivers established to the millimeter level on baselines
of, say, up to 50 kilometers in length. Centimeter-level accuracy is available on base-
lines of much longer length, but uncertainties introduced by the Earth's atmosphere
and the satellite positions cause increasing errors with increasing baseline length.
As has already been described in Chapter 2, in addition to the positional informa-
tion, the user's clock can be set relative to the GPS satellite clocks to accuracies as
good as ± 10 nanoseconds or better, thus providing a time-transfer mechanism of
extremely high accuracy.
For the purposes of Earth rotation, one must assume that three or more satellite-
tracking stations are in use at known locations. It is then possible to determine the
position of the satellites with respect to the ground stations, as well as the orienta-
tion of the Earth with respect to the "inertia]" coordinate system established by the
satellites. In practice, such tracking will be done by the "control segment," Dept.
of Defense tracking stations responsible for operating the satellite, other govern-
ment networks (e.g., that operated by the U.S. National Geodetic Survey (NGS) at
U.S. IRIS VLBI stations), or private groups. Any of these data are now partially
usable for Earth rotation determination, and should be capable of providing high-
accuracy ERP values once the entire GPS satellite system is operational. Abbot et
al. (1988) have already used data obtained in 1985 to show that errors could be
kept to ± 40 mas in xp, ± 70 mas in yp, and + 0. 9rns in UT1-UTC over 5 days of
measurement. Zelensky et al. (1990), in a covariance analysis of observations of 18
Block II satellites, show that accuracies in the range of 0.2 to 1.0 mas for all three
ERP components will be possible. Paquet and Louis (1990) in a simulation (with
18 satellites and 10 tracking stations making + 10 cm accuracy measurements) show
possible accuracies in the range of less than ± 3 mas for 24 hours and ± 3-5 mas for
6 hours of observation for pole position, and ± 0. 15-0.30 ms for UT1-UTC.
(For another general overview of the GPS, see Archinal and Mueller (1986).
More detailed information is also available in Chapter 3 and in Remondi (1984),
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 261
Goad (1985), Applied Research Laboratories (1986), Institute of Navigation (1980,
j984), and Physical Sciences Laboratory (1989).)
A similar system, known as the Global Navigation Satellite System (GLONASS),
is also under development by the former republics of the USSR. It consisted of 10 to
12 satellites by the end of 1990, and 24 satellites sometime in the period 1991-1995.
Its primary purpose will be for worldwide aircraft navigation, but it is also planned
to be used by naval and fishing fleets as well. The system is in reality very closely
patterned after the GPS, with the primary exception that each satellite will trans-
mit different L-band frequencies (rather than just using different codes as the GPS
satellites do). The receiver hardware, therefore, will have some differences from that
of GPS receivers; however, some current receivers are, and it is likely many future
ones will be, able to receive and process signals from either system. Some additional
details on this system are presented in Anodina and Prilepin (1989).
4.442 Laser-Ring Gyroscopes Another more exotic method for determining vari-
ations in the Earth's rotation has also been proposed. If a laser beam is split and
sent in opposite directions around a given path on the Earth's surface, the light can
be made to interfere with itself upon meeting again. Changes in this interference
pattern will occur because of changes in the Doppler shift of the light traversing
the two different paths, due to variations in the Earth's rotation. Such a device is
known as a "laser ring" or "Sagnac laser" gyroscope (Soffel et al., 1988). Although
theoretically it might be able to provide usable Earth rotation data, in practice the
stability of the platform on which the laser light is directed seems to be a limiting
factor, because of changes in light paths due to vibrations, microearthquakes, etc.
Such a device (covering a 58-m2 area) is currently undergoing testing and upgrad-
ing, with reported accuracies of 10_s over averaging times on the order of a day,
and a predicted theoretical accuracy of a few parts in 10~10 with averaging times
of a few seconds (Rotge et al., 1985).
4.443 Superfluid Helium Interference Gyroscopes The use of superfluid helium
interference gyroscopes (SHIG) has just recently been proposed for the determi-
nation of Earth rotation and polar motion (Packard, 1988). It involves the use of
a small (less than 1-meter diameter) horizontal torus of superfluid helium (3He or
He), in which a blockage with a pinhole is positioned. When this torus undergoes
rotation (for example, because of the rotation of the Earth), it should be possi-
ble to measure the phase difference or phase slips across the blockage (Avenel and
Varoguaux, 1988). This in turn is proportional to the rotational velocity of the
torus and therefore the Earth, and could be measured to 1 part in 109 over a few
seconds. Of course technical problems still exist in building such a device, such as
262 EXPLANATORY SUPPLEMENT
(a) building a pinhole diaphragm of very small aperture (0.1 fi holes are needed),
and (b) stabilizing the instrument or determining its tilt with respect to the local
vertical (i.e., its connection to the Earth) with the accuracy needed (changes of
10-9 meters in tilt would be significant across a 1-meter instrument). If they can
successfully be built, however, two or more of this type of instrument, as with the
laser-ring gyroscopes, would provide high-speed determinations of LOD and polar
motion independent from astronomical observations.
4.45 International Services
On 1988 January 1, the International Earth Rotation Service (IERS) began op-
eration, under the auspices of the IAU and the IUGG. It replaced the Bureau
International de l'Heure (BIH) and the International Polar Motion Service (IPMS)
as the organization responsible for determining and predicting the orientation of the
Earth. The IERS consists of a board of directors that oversees a "central bureau"
and "coordinating centers" for each observational technique and "sub-bureaus" for
special purposes. Further, each technique may have several "computational centers"
actually responsible for processing data and ERP series that are then collected,
used, and distributed by the Central Bureau.
The IERS publishes two different primary bulletins for the dissemination of
Earth-rotation data as determined from the combination of data from different
observational techniques. The IERS Bulletin A provides information weekly on
the rapid determination of the Earth's rotation, as well as predictions for future
rotation. The IERS Bulletin B provides "final" smoothed data monthly on the
rotation of the Earth. Of course, other organizations both within and outside the
IERS also provide other ERP series, usually based on observational data of a given
type.
The IERS is also responsible for deciding when a step in UTC is necessary in
order to keep it within 0.7 s of UT1. These steps, commonly known as leap seconds,
may be positive or negative, are of 1 second duration, are announced at least 8
weeks ahead of time, and may be inserted at the end of any month (preferably,
and in practice, at the end of June or December) (BIH, 1972, pp. A-ll to A-14).
Since their introduction in 1972, these steps have always been positive, and have
been required from every 6 months to 2 years. A table of these steps is given in the
current Astronomical Almanac, p. K9, as differences of UTC from TAI.
4.451 IERS Bulletin A The IERS Bulletin A is published by the National Earth
Orientation Service (NEOS), a joint activity of the U.S. Naval Observatory and the
4 1 t£Rrestrial coordinates and the rotation of the earth 263
National Geodetic Survey of the United States. The NEOS acts as a sub-bureau
0f the IERS. The IERS Bulletin A is issued by 0 hours UT, Friday, of each week,
and contains Earth rotation parameters determined from the combination of other
recently determined ERP series, predictions of Earth rotation parameters daily for
Up to 90 days in the future, predictions of Earth rotation parameters monthly up
to one year in the future, and equations for predictions even further into the future.
Other miscellaneous data and information are also available, such as the ERP de-
termined from observations being used in the combined solution, plots of current
polar motion and UT1— UTC values, values of TAI— UTC, DUT1, announcements
of leap seconds, etc. For the combined ERP series, data given includes for each
day: the calendar date and modified Julian date (MJD); the values of xp, yp, and
their expected accuracies in arcseconds; and the values of UTl^UTC and their
expected accuracies in seconds of time. Values are usually given for approximately
a week's worth of data. When available, new values from the IERS Bulletin B are
also published. Predictions are given for each day for 90 days past the end of the
combined data, consisting of the calendar date and MJD, values of xp and yp in
arcseconds, UT1— UTC in seconds, and a table giving estimated accuracies of these
values. Equations are also provided in order to extend these predictions even fur-
ther. Similar predictions are given for the first day of each month for one year in
the future. Predicted celestial pole offsets (corrections to Aip and Ae) are also given
for each day for 90 days in the future.
The combined series being published is actually formed at the U.S. Naval Ob-
servatory, using ERP series based on LLR, SLR to LAGEOS, and VLBI data. The
individual series in use are provided by the Department of Astronomy at the Univer-
sity of Texas (UTX) using LLR data; the Center for Space Research at UTX and the
Technical University of Delft for SLR to LAGEOS data; and the National Geodetic
Survey and the U.S. Naval Observatory for VLBI data. Each series has systematic
errors removed from each of its available components (xp, yp, and UT1— UTC). A
spline curve fitted through each of the components is used to determine the one- day
values for each component. Accuracy estimates for each input series are estimated
periodically via intercomparisons of the series. The long-term biases and trends are
adjusted on an annual level to those of the series published in the IERS Bulletin
B. The predictions for xp and yp are done from a least-squares solution using the
last two years of data. The predictions for UT1-UTC are computed using a model
based on the last 365 days of data. A complete description of the combination and
prediction technique is given in NEOS (1989) and McCarthy and Luzum (1990).
All of the IERS Bulletin A information is also available on the "Automated
Data Service" of the U.S. Naval Observatory, and from the Arpanet, BITNET, In-
ternet, G.E. Mark III, and SPAN electronic networks. For further information, or to
b'1 placed on the mailing list for the IERS Bulletin A, contact: National Earth Ori-
entation Service, Earth Orientation Parameters Division, U.S. Naval Observatory,
Washington, D.C. 20392.
264 EXPLANATORY SUPPLEMENT
4.452 IERS Bulletin B The IERS Bulletin B is published by the Central Bureau
of the IERS, located at the Paris Observatory in Paris, France. It is issued during the
first week of each month, and contains final ERP values from a combination solution
of a month's or more ERP series data, more recent ERP values and predictions as
available from the IERS Bulletin A, along with information on the length of day
(LOD) and ERP series data used in the combined solution. The final values, given
for every 5 days, include the calendar date and MJD, smoothed values of xp and
yp in arcseconds, UT1R-UTC and UT1R-TAI in seconds, and raw values of xp
and yp in arcseconds, and UT1-UTC in seconds. Following this are some of the
rapid service and prediction values from IERS Bulletin A in the same format. These
values are also interpolated daily and given with the calendar date and MJD, x
and yp in arcseconds, UTl-UTC, UTl-TAI, UTl-UTlR and the excess length
of day ("D") in seconds. Smoothed values of "D" and the rotational velocity of the
Earth in microradians/second are also given for every 5 days.
The final values ("raw" above) are computed for every 5 days, using data from
various observational ERP series as input. The input series are the same as those
listed for the IERS Bulletin A, with, in addition, series obtained from JPL VLBI
observations also being used. The values are then smoothed using a "Vondrak"
filter (Vondrak, 1969). The values of UTl-UTC are also corrected for variations
due to zonal Earth tides with periods up to 35 days, giving UTlR-UTC. The
values are referred to the IERS Terrestrial System, which is identical to the earlier
BIH Terrestrial System, with the BIH origin having been adjusted to the CIO in
1967 and maintained independently since then. This series is also a direct extension
of the BIH Circular D series, which ended on December 31, 1987.
All of this information is also available from the G.E. Mark III and the EARN/
BITNET electronic networks. For further information, or to be placed on the mail-
ing list for the IERS Bulletin B, contact: Central Bureau of the IERS (IERS/CB),
Observatoire de Paris, 61, Avenue de l'Observatoire, 75014 Paris, France.
4.453 Other Publications of the IERS Other publications of the IERS include:
Special Bulletin C — Issued irregularly to announce upcoming steps (leap sec-
onds) in UTC.
Special Bulletin D — Issued irregularly to announce upcoming changes to the
value of DUT1 to be transmitted with time signals.
Annual Report of the IERS — Issued annually from 1989 June on, reporting on
various solutions for Earth orientation parameters and terrestrial and celestial
reference systems.
Technical Notes of the IERS — Issued irregularly as reports and complemen-
tary information of relevance to the work of the IERS on Earth rotation and
related reference systems.
/ TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 265
For further information, contact the Central Bureau of the IERS at the address
given in Section 4.452.
4 5 DETERMINATION OF PAST VARIATIONS IN LENGTH OF DAY
AND THE POSITION OF THE POLE
The determination of historical variations in the rotational rate of the Earth has
been a topic of discussion since the latter part of the nineteenth century, when such
variations were first suspected. Most of the research in this area has centered on the
use of ancient astronomical observations of eclipses and telescopic observations of
lunar occultations, in order to determine At, the difference between Earth rotational
time (UT1) and dynamical time (TDT or formerly ET). The time derivative of this
value is the length of day (LOD). A summary of recent work in this area will be
presented along with some current results.
In addition, available results on historical variations in polar motion are pre-
sented. However, since these variations do not result in a cumulative effect as they
do with UT1, older measurements of these variations are based entirely on precise
telescopic observations and so can only be extended back to the mid-nineteenth
century with any degree of continuity.
4.51 Historical Variations in UT1 and Length of Day
Since studies by Newcomb (1882, p. 465) indicated that measurements of the tran-
sits of Mercury, the Moon, and Jupiter's satellites all showed similar variations
possibly related to the Earth's rotation, many studies and papers have been un-
dertaken to determine more accurately these variations over time. However, these
studies have been complicated by their dependence on the position of the Moon
(during eclipses or occultations), and hence on whether there has ever been any
change in the acceleration of the Moon's mean motion (h) (and effectively, then,
any change in the rate of ET determinations based on the Moon's motion). Many
studies of this problem that have been undertaken up until the 1970s attempted
to solve simultaneously for h and At. An example of this type of approach is that
by Newton (1970). However, h can be assumed constant, an assumption that seems
to be supported by modern astronomical observations (including radar and LLR
measurements) of the Moon. Many studies during the 1970s and 1980s have been
undertaken with this assumption as their basis. The most complete current ref-
erences on historical UT1 and LOD variations (with n assumed constant) are by
Stephenson and Morrison (1984), and McCarthy and Babcock (1986) from which
much of the material presented here is derived. The results presented in both of these
references agree well, with differences due only to slightly different data sources and
266 EXPLANATORY SUPPLEMENT
different degrees of smoothing before 1861 (McCarthy, personal communication).
Some newer results are also available in Stephenson and Lieske (1987).
The results can conveniently be considered by examining various historical pe-
riods separately, based upon the type of observations being considered. The specific
periods to be mentioned here will include (working backward) 1980-present, 1955-
1980, 1620-1954, 700-1600, and 700 B.C.-A.D. 200. Observations also exist for the
period A.D. 200-700, but have not yet become available (in published form). For
example Hilton et al. (1989) report At values determined from 58 occupations of
planets by the Moon observed by the Chinese in the period 68 B.C. to A.D. 575.
Extended details of the methods for the determinations of UTl and LOD will
not be presented here. For further information on the methods used when consid-
ering lunar occultation data, Morrison (1979b) may be consulted. Descriptions of
how eclipse data were used are presented in Stephenson and Morrison (1984) along
with many additional references on data sources. Other chapters here may also
be consulted. Current methods for observing UTl (since 1980) are fully described
earlier (in Section 4.5).
Observations of UTl from 1955 to 1980 are unique in that they can be refer-
enced to TDT, due to the advent of the first atomic time standards in the 1950s.
However, since mostly optical techniques and then developing space-based tech-
niques were in use, the accuracy levels were an order of magnitude lower than at
the present time. Determinations by the BIH provide the best results from 1963-
1980 (originally in the form of TAI-UTl, with TAI different from TDT only by
an additive constant to TAI of 32.184s). McCarthy and Babcock (1984) present
the most accurate available results during the period 1956-1962, based on USNO
Photographic Zenith Tube (PZT) observations of UTl, as referenced to the USNO
atomic timescale A.l (which is different from TDT only by an additive constant to
A.l of 32.150s).
The period of 1620 to 1955 covers substantially the period where telescopic
observations may be used in determinations of UTl. Lunar occultations alone pro-
vide fairly precise values from 1861 to 1954 as described by Morrison (1979a). The
sparseness of such observations before that time results in some difficulties in re-
covering UTl. Results from occultations and now also eclipses are fitted with cubic
splines in order to obtain a continuous series of values. Beyond Stephenson and
Morrison's 1984 treatment, part of this period (1672-1806) has been reconsidered
with newly available solar eclipse timings by Stephenson and Lieske (1987).
The "medieval period" of 700-1600 refers to the general period from timed
Arabian eclipse observations and other untimed observations until the invention of
the telescope. During this period, a single "curve" of values for At may no longer
be estimated. One may only set limits on the values at sporadic epochs during
the period. The timed Arabian observations actually ran from approximately A.D.
800 to 1000, with the timings made by measuring the altitude of the Moon or a
i TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
Length of day — 86400 seconds
267
6
4
I
O
8 -0
m
3 .2
I
-4
-6
r
-
/A
jr f "
V '
" 1
— i
1600
1700
1800
Year
1900
2000
Figure 4.5L1
Length of day from 1656 to IS
clock star. The untimed observations were made largely from Europe, usually in
nonastronomical works.
"Ancient" observations during the period circa 700 B.C. to A.D. 200 are of a
rather limited number, but in some cases are quite useful. Stephenson and Morrison
(1984) have been able to use (a) 5 Chinese sightings of total and near total eclipses
between 198 B.C. and A.D. 120); (b) a Babylonian observation of atotal solar eclipse
in 136 B.C.; (c) "timed" (possibly with water clocks) Babylonian observations of
lunar eclipses; and (d) 20 Babylonian records indicating that the Moon rose or set
while eclipsed.
As a summary of the preceding results, graphs of At are presented in Figures
4.51.1-4.51.3. Figure 4.51.1. presents a curve of At vs time for the recent historical
period of 1656 to the present, after McCarthy and Babcock (1985) with recent
values from the IERS Rapid Service Sub-Bureau. The determinations of At by
Stephenson and Morrison (1984) from isolated observations are presented for the
Period A.D. 700 to 1600 in Figure 4.51.2, and for the period 700 B.C. to A.D. 200 in
Figure 4.51.3. Finally, their logarithmic plot of Ar is presented for 700 B.C. to A.D.
1980 in Figure 4.51.4. The current Astronomical Almanac also contains a table of
At values for the period 1620.0 to the present.
T
timed lunar
+ and solar
eclipses.
T untirned solar
eclipse
f observations
800
1000
1200
1400
1(500
Figure 4.51.2
At from a.d. 700 to 1600
o
St
<
• timed lunar
0 observations
untimed
solar eclipse
observations
T untimed lunar
1 eclipse
I observations
• 0
; coo
Si !
r
i i
-600
-400
-200
year
200
Figure 4.51.3
At from 700 B.c, to A.d. 200
TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
269
-1000 0 1000 2000
year
Figure 4.51.4
log At from 700 B.C. to A.D. 1980 (30 seconds have been added to each value to make them all positive.
The continuous line shows the computed tidal deceleration of the Earth.)
4.52 Historical Variations in Polar Motion
The advent of the International Latitude Service in the 1890s brought about moni-
toring of polar motion that has continued to this day. However, except for sporadic
periods during the nineteenth century, little data on polar motion exists before that
time.
Extremely high-precision polar-motion data, as with UT1 data, exists only from
1980 on, with the introduction of the space-based geodetic techniques discussed in
Section 4.4. The recent motion of the pole is presented in Figure 4.52.1, using values
Provided by the IERS Rapid Service Sub-Bureau.
High-accuracy polar-motion data, based on large numbers of optical astrometric
instruments became available just prior to the formation of the International Polar
Motion Service in 1962. Doppler satellite tracking data also began providing polar-
motion information in the late 1960s, and was in use by the BIH by the early 1970s.
270
EXPLANATORY SUPPLEMENT
Polar motion
-.4
-.2-
ti
■4
.0-
.2-
I 1 meter
v 1990
I 1 (U981 1
1 1
1
.5 .3
y (sec. of arc)
-.1
Figure 4.52.1
Polar motion, 1980 September 28 to 1990 July 27 (triangles mark the beginning of
each year)
This type of data still continues to be available until the present day, but has been
superseded in accuracy by the newer techniques.
The stations of the ILS provided monitoring of polar motion from September
of 1899 until 1987. A complete and rigorous rereduction of all of the ILS data
from 1899.9 until 1979.0 is described and reported by Yumi and Yokoyama (1980).
Figure 4.52.2, showing the X and Y values versus time during that period, is taken
from Figure 8 of that publication. Fedorov et ai (1972) also present polar-motion
results from 1890.0 to 1969.0.
Before 1900, data that can be used for polar-motion determination are quite
sporadic. All such data are based on short observation campaigns, usually at one
station, and although some observers were in fact looking for polar motion, most
of the observations were transit-circle or visual zenith instrument observations be-
ing used to determine star positions and/or time. Polar-motion data are available
from the previously mentioned work by Fedorov et al. from 1890.0 to 1900.0. Cov-
ering a more lengthy period during the nineteenth century, specifically during the
period 1846-1891.5, is the study by Rykhlova (1969). Figure 4.52.3 shows the polar-
motion results during that period. Surprisingly, one of the other few discussions of
polar-motion data before 1900 is that of Chandler himself (1891, 1892). Using data
0.6
ILS polar motion in y
Figure 4.52.2
X and Y components of polar motion from the ILS. 1 899.9 to 1979.0
dating from 1863 to 1885 from various observatories, he demonstrated decisively
the existence of the motion now known as the Chandler wobble. Chandler actually
considered data over the entire period 1726 (from Bradley) to 1890, which indicated
large changes were occurring in the length of the Chandler period. These changes
are usually considered to be artifacts of the sparse low-precision data in use.
ml) U tS SI St S! S3 St sTsl ST k S3 SO Si K S3 « SS ii S7 Si SI W 71 ?2 V » ff » « 7t ?! SS 81 SI t3 ft Si 16 17 X ti w
Figure 4.52.3
X and Y components of polar motion, 1 846.0 to 1 891 .5
272
EXPLANATORY SUPPLEMENT
4.6 REFERENCES
Abbot, R.I., King, R.W., Bock, Y., and Counselman III, C.C. (1988). "Earth Rotation
from Radio Interferometric Tracking of GPS Satellites," in Babcock and Wilkins,
1988, 209-213.
Agnew, D.C. (1983). "Conservation of Mass in Tidal Loading Computations" Geophys. J.
Roy. Astron. Soc. 72, 321-325,
Alley, CO. (1983). "L aser Ranging to Retro-Reflectors on the Moon as a Test of Theo-
ries of Gravity" in Quantum Optics, Experimental Gravity, and Measurement Theory
(Plenum, New York, NY), 431.
Applied Research Laboratories (1986). Proceedings of the Fourth International Symposium
on Satellite Positioning (ARL, University of Texas, Austin).
Anodina, T.G. and Prilepin, M.T. (1989). "The GLONASS System" in Physical Science
Laboratory, 1989, 1, 13-18.
Archinal, B. and Mueller, I.I. (1986). "GPS An Overview" Professional Surveyors, no. 3,
7 11.
Arias, E.F., Feissel, M., and J.-F. Lestrade (1988). "An Extragalactic Celestial Reference
Frame Consistent with the BIH Terrestrial System (1987)," in BIH 1988, D-113 to
D-121.
Armstrong, N. (1977). Foreword to Mulholland 1977, xiii-xiv.
Arnold, K. (1974), "Geodetic Aspects of Laser Distance Measurements to the Moon and
Radio-Interference Measurements to Quasars" Gerlands Beitr. Geophysik 83, no. 4,
249-269.
The Astronomical Almanac (1990). (U.S. Government Printing Office, Washington, DC),
Avenel, O., and Varoquaux, E. (1988). "Josephson Effect and Quantum Phase Slippage
in Superfluids" Physical Review Letters 60, no. 5, 416-419.
Babcock, A.K. and Wilkins, G., eds. (1988). The Earth's Rotation and Reference Frames
for Geodesy and Geodynamics IAU Symposium 128 (Khiwer, Dordrecht, Holland),
BIH (1973). BIH Annual Report for 1972 (BIH, Paris Observatory, Paris, France).
BIH (1989). BIH Annual Report for 1988 (BIH, Paris Observatory, Paris, France).
Bock, Y. (1980). "A VLBI Variance-Covariance Analysis Interactive Computer Program"
Dept. of Geodetic Science Report 298 (Ohio State University, Columbus, OH).
Bomford, G. (1971). Geodesy, 3rd ed. (Oxford University Press, London, UK).
Borkowski, K.M. (1989). "Accurate Algorithms to Transform Geocentric to Geodetic Co-
ordinates" Bulletin Geodesique 63, no. 1, 50-56,
Boucher, C. and Altamimi, Z. (1989). "The Initial IERS Terrestrial Reference Frame"
IERS Technical Note 1 (IERS, Paris, France).
Brosche, P., Seiler, U., Suendermann, J., and Wuensch, J. (1989). "Periodic Changes in
Earth's Rotation due to Oceanic Tides" Astronomy and Astrophysics 220, 318-320.
Carter, W.E. and Robertson, D.S. (1984). "IRIS Earth Rotation and Polar Motion Re-
sults" in Proc. Int. Symposium Space Techniques for Geodynamics Vol. 1 J. Somogyi
and C. Reigber, eds., (Hungarian Academy of Sciences, Sopron, Hungary), pp. 214-
222.
Carter, W.E., Robertson, D.S., Pettey, J.E., Tapley, B.D., Shutz, B.E., Eanes, R.J., and
Lufeng, M. (1984). "Variations in the Rotation of the Earth" Science 224, 957-961.
Cartwright, D.E. and Edden, A.C. (1973). "Corrected Tables of Tidal Harmonics" Geo-
phys. J. Roy. Astron. Soc. 33, 253-264.
TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
273
Cartwright, D.E. and Tayler, R.J. (1971). "New Computations of the Tide-Generating
Potential" Geophys. J. Roy. Astron. Soc, 23, 45-74.
Chandler, S.C. (1891). "On the Variation of Latitude" The Astronomical Journal 11, 83,
Chandler, S.C. (1892). "On the Variation of Latitude" The Astronomical Journal 12, 17.
Chapman, S. and Lindzen, R. (1970). Atmospheric Tides (D. Reidel, Dordrecht, Holland).
Clark, T.A., Corey, B.E., Davis, J.L., Elgered, G., Herring, T.A., Hinteregger, H.F.,
Knight, C.A., Levine, J.I., Lundqvist, G., Ma, C, Nesman, E.F., Phillips, R.B.,
Rogers, A.E.E., Ronnang, B.O., Ryan, J.W., Schupler, B.R., Shaffer, D.B., Shapiro,
I.I., Vandenberg, N.R., Webber, J.C., and Whitney, A.R. (1985). "Precision Geodesy
Using the Mark-Ill Very-Long-Baseline Interferometer System" IEEE Trans, on
Geosci. and Rem. Sens. GE-23, 4, 438-449.
Coates, R.J., Frey, H., Mead, G.D., and Bosworth, J.M. (1985). "Space-Age Geodesy: The
NASA Crustal Dynamics Project" IEEE Trans, on Geosci. and Rem. Sens. GE-23,
4, 360-368.
Cohen, S.C, King, R.W., Kolenkiewicz, R,, Rosen, R.D., and Schutz, B.E., eds. (1985).
"LAGEOS Scientific Results" special issue of Journal of Geophysical Research 90,
Bll, 9215-9438.
Degnan, J. (1985). "Satellite Laser Ranging: Current Status and Future Prospects" IEEE
Trans, on Geosci. and Rem. Sens. GE-23, 4, 398-413.
DeMets, C, Gordon, R.G., Argus, D.F., and Stein, S. (1990), "Current Plate Motions"
Geophys. J. Int. 101, 425-478.
Department of the Army (1958). Universal Polar Stereographic Grid TM 5-241-9 (De-
partment of the Army, Washington, DC).
Department of the Army (1973). Universal Transverse Mercator Grid TM 5-241-8 (De-
partment of the Army, Washington, DC).
DMA (1987). Supplement to Department of Defense World Geodetic System 1984 Tech-
nical Report DMA TR 8350. 2-A (Defense Mapping Agency, U.S. Naval Observatory,
Washington, DC).
Doodson (1921). Proc. R. Soc. A. 100, 305-329.
Eanes, R.J., Schutz, B., and Tapley, B. (1983). "Earth and Ocean Tide Effects on LAGEOS
and Starlette" Proceedings of the Ninth International Symposium on Earth Tides (E.
Schweizerbartsche Verlag Buchhandtung, Stuttgart).
Farrell, W.E. (1972). "Deformation of the Earth by Surface Loads" Rev. Geophys. Space
Phys. 10, 761-797.
Fedorov, E.P., Korsun, A. A., Mayor, S.P., Panchenko, N.I., Taradii, V.K., and Yatskiv,
Y.S. (1972). The Motion of the Pole of the Earth from 1890.0 to 1969.0 (Ukrainian
Academy of Sciences, Naukova Dumka, Kiev, USSR).
Gilbert, F. and Dziewonski, A.M. (1975). "An Application of Normal Mode Theory to
the Retrieval of Structural Parameters and Space Mechanisms from Seismic Spectra"
Phil Trans. R. Soc. of Land. A278, 187-269.
Goad, C.C. (1980). "Gravimetric Tidal Loading Computed from Integrated Green's Func-
tions" J. Geophys. Res. 85, 2679-2683.
Goad, Clyde (ed.) (1985). Proceedings of the First International Symposium on Precise Po-
sitioning with the Global Positioning System (National Geodetic Information Center,
NOAA, Rockville, MD).
GSFC (1976). Mathematical Theory of the Goddard Trajectory Determination System X-
582-76-77 (GSFC, Greenbelt, MD),
Heiskanen, W. and Moritz, H. (1967). Physical Geodesy (W.H. Freeman, San Francisco,
CA).
274
EXPLANATORY SUPPLEMENT
Hilton, J.L., Seidelmann, P.K., and L. Ciyuan (1989). "Chinese Records of Lunar Occul-
tations of the Planets" Bull, of Am. Ast. Soc. 21, no. 4, 1153. Abstract only.
IAU (1983). "Proceedings of the Eighteenth General Assembly Patras 1982" Trans of the
I A V 18B (Reidel, Dordrecht, Holland).
IERS (1988). IERS International Earth Rotation Service (U.S. Naval Observatory, Wash-
ington, DC).
IERS (1989). An Explanatory Supplement to the IERS Bulletins A and B {IERS, U.S.
Naval Observatory, Washington, DC).
Institute of Navigation (1980). Global Positioning System (The Institute of Navigation
Washington, DC).
Institute of Navigation (1984). Global Positioning System Volume II (The Institute of
Navigation, Washington, DC).
Kaula, W.M. (1966). Theory of Satellite Geodesy (Blaisdell, Waltham, MA).
Kaula W.M. (1975). "Absolute Plate Motions by Boundary Velocity Minimizations" J.
Geophys. Res. 80, 244-248,
Klepczynski, W.J. (1982). "Coordinator's Report: Connected-Elements Radio Interferom-
etry" in Project MERIT: Report on the Short Campaign G.A. Wilkins and M. Feissel,
eds. (Royal Greenwich Observatory, Herstmonceux, UK), 34, 77.
Kouba, J. (1983). "A Review of Geodetic and Geodynamic Satellite Doppler Positioning"
Rev. Geophys. Space Phys. 21, 27-40.
Kovalevsky, J. and Mueller, I. (1981). "Comments on Conventional Terrestrial and Quasi-
Inertial Reference Systems" in Reference Coordinate Systems for Earth Dynamics E.
Gaposchkin and B. Kolaczek, eds, (Reidel, Dordrecht, Holland), pp. 375-384.
Kovalevsky, J., Mueller, I., and Kolaczek, B., eds. (1989). Reference Frames in Astronomy
and Geophysics (Kluwer, Dordrecht, Holland).
Kumar, M. (1972). "Coordinate Transformation by Minimizing Correlations between Pa-
rameters" Dept. of Geodetic Science Report 184 (Ohio State University, Columbus,
OH).
Lambeck, K. (1971). "Determination of the Earth's Pole of Rotation from Laser Range
Observations to Satellites" Bulletin Geodesique 101, 263-280.
Lambeck, K. (1980). The Earth's Variable Rotation (Cambridge University Press, Cam-
bridge, England).
Larden, D.R. (1982). "Monitoring the Earth's Rotation by Lunar Laser Ranging" Reports
from the School of Surveying UNISURV S-20 (Univ. of New South Wales, Kensington,
Australia).
Ma, C. (1978). Very Long Baseline Interferometry Applied to Polar Motion, Relativity and
Geodesy TM 79582 (NASA GSFC, Greenbelt, MD).
Markowitz, W., Hall, R.G., Essen, L., and Parry, J.V.L. (1958). "Frequency of Cesium in
Terms of Ephemeris Time" Physical Review Letters, I, no. 3.
Markowitz, W. (1960). "The Photographic Zenith Tube and the Dual Rate Moon Camera"
Stars and Stellar Systems G.P. Kuiper and B.M. Middlehurst, eds., I (University of
Chicago Press, Chicago, IL).
Marsh, J.G., Lerch, F.J., Putney, B.H., Christodoulidis, D.C., Smith, D.E., Felstentreger,
T.L., Sanchez, B.V., Klosko, S.M., Pavlis, E.C., Martin, T.V., Robbing, J.W., William-
son, R.G., Colombo, O.L., Rowlands, D.D., Eddy, W.F., Chandler, N.L., Rachlin,
K.E., Patel, G.B., Bhati, S., and Chinn, D.S. (1988). "A New Gravitational Model
for the Earth from Satellite Tracking Data: GEM-T1" J. Geophys. Res. 93, 6169-6215.
Martin, T.V., Oh, I.H., Eddy, W.F., and Kogut, J. A. (1976). Volume I GEODYN System
Description (EG&G/WASC, Riverdale, MD).
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH
275
Matsakis D.N., Jostles. F.J., Angerhofer, P.E., Florkowski, D.R., McCarthy, D.D, Ji-
ayan' X., and Yunlou, P, (1986). "The Green Bank Interferometer as a Tool for the
Measurement of Earth Orientation Parameters" The Astronomical Journal, 91, no. 6,
1463-1473.
McCarthy D D , Angerhofer, P.E, and Florkowski, D.R. (1985). "Connected-Element
Interferometer Results in Project MERIT" in Mueller, 1985, 1, 341-350.
McCarthy, D.D. and Babcock, A.K. (1986), "The Length of Day since 1656" Physics of
the Earth and Planetary Interiors 44, 281-292.
McCarthy, D.D., ed. (1989). "IERS Standards (1989)" IERS Technical Note 3 (Central
Bureau of the IERS, Observatoire de Paris, Paris, France).
McCarthy, D.D. (1990). "High Frequency Variations In Earth Rotation" in McCarthy et
id., 1990.
McCarthy, D.D. and Luzum, B. (1990). "Combination of Precise Observations of the
Orientation of the Earth" Submitted to Bulletin Geodesique.
McCarthy, D.D., Paquet, P., and Carter, W., eds. (1990). Proceedings of the IUGG Sym-
posium "Variations in Earth Rotation" (American Geophysical Union, Washington,
DC).
McLintock, D.N. (1980). "Very Long Baseline Interferometry and Geodetic Applications
PhD dissertation (Univ. of Nottingham, Dept. of Civil Engineering, UK).
Minster, J.B. and Jordan, T.H. (1978). "Present-Day Plate Motions" J. Geophys. Res. 83,
Bll, 5331-5354.
Mitchell, H.C. and Simmons, L.G. (1945). The State Plane Coordinate Systems {A Manual
for Surveyors) Special Publication 235 (National Geodetic Survey, Rockville, MD).
Moffit, F. and Bouchard, H. (1975). Surveying, 6th ed. (Intext Educational Publishers,
New York, NY).
Molodenskii, M.S., Eremeev, V.F., Yurkina, M.I. (1962). Methods for Study of the External
Gravitational Field and Figure of the Earth (Israel Program for Scientific Translations,
Jerusalem) .
Moritz, H. (1980). Advanced Physical Geodesy (Abacus Press, Tunbridge Wells, Kent, UK).
Moritz, H. and Mueller, I.I. (1987). Earth Rotation: Theory and Observation (Ungar/Con-
tinuum, New York, NY).
Morrison, L.V. (1979a). "Re- determination of the Decade Fluctuations in the Rotation of
the Earth in the period 1861-1978" Geophys. J.R. Astr. Soc. 58, 349-360.
Morrison, L.V. (1979b). "An Analysis of Lunar Occultations in the Years 1943-1974 for
Corrections to the Constants in Brown's Theory, the Right Ascension System of the
FK4, and Watts' Lunar-profile Datum" Man. Not. R. Astr. Soc. 187, 41-82.
Mueller, I.I., and Rockie, J.D. (1966). Gravimetric and Celestial Geodesy: A Glossary of
Terms (Ungar, New York, NY).
Mueller, I.I. (1969). Spherical and Practical Astronomy as Applied to Geodesy (Ungar,
New York, NY).
Mueller, I.I., Kumar, M, Reilly, J.P., Saxena, N., and Soler, T. (1973). "Global Satellite
Triangulation and Trilateration for the National Geodetic Satellite Program (Solu-
tions WN 12, 14, and 16)" Department of Geodetic Science Report No. 199 (The Ohio
State University, Columbus, OH).
Mueller, I.I., ed. (1985). Proceedings of the International Conference on Earth Rotation
and the Terrestrial Reference Frame (Dept. of Geod. Sci. and Surveying, The Ohio
State University, Columbus, OH).
Mueller, I.I. (1988). "Reference Coordinate Systems: An Update" in Theory of Satellite
Geodesy and Gravity Field Determination R. Rummel and F. Sanso, eds. (Springer
Verlag, Heidelberg).
276
EXPLANATORY SUPPLEMENT
Mulholland, J.D., ed. (1977). Scientific Applications of Lunar Laser Ranging {Reidel, Dor-
drecht, Holland).
Munk, W.H. and MacDonald, G.J.F. (1960). The Rotation of the Earth: A Geophysical
Discussion (Cambridge Univ. Press, Cambridge, UK).
Newton, R.R. (1970). Ancient Astronomical Observations and the Accelerations of the
Earth and Moon (The Johns Hopkins Press, Baltimore, MD).
Newcomb, S. (1882). "Discussion and Results of Observations on Transits of Mercury,
from 1677 to 1881" Astr. Pap. Am. Eph. 1, part 6, 363-487.
NGS (1985). "Geodetic Software" Information flyer 85-10 (NGS, Rockville, MD 20852).
Reprinted January 1987.
Packard, R.E. (1988). "Principles of a Superfluid Helium Interference Gyroscope: A New-
Tool to Study the Earth's Rotation" EOS 69, no. 44, 1153. Abstract only.
Pagiatakis, S.D., Langley, R.B., and Vanicek, P. (1984). "Ocean Tide Loading: A Global
Model for the Analysis of VLBI Observations" in P roc. of the 3rd International Sym-
posium on the Use of Artificial Satellites for Geodesy and Geodynamics Ermioni,
Greece, Sept. 1982, G. Veis, ed. (National Techn. Univ., Athens), 328-340.
Paquet, P. and Louis, L. (1990). "Simulations to Recover Earth Rotation Parameters with
GPS System" in McCarthy et d,, 1990.
Parker, W. and Bartholomew, R. (1989). "Issues in Automating the Military Grid Refer-
ence System" in Physical Science Laboratory, 1989, 1, 303-913.
Physical Science Laboratory (1989). Proceedings of the Fifth International Geodetic Sym-
posium on Satellite Positioning, 2 vols. (New Mexico State University, Las Crnces,
NM).
Rabbel, W. and Schuh, H. (1986). "The Influence of Atmospheric Loading on VLBI Ex-
periments" J. Geophysics 59, 164-170.
Rapp, R.H, (1980). Geometric Geodesy Volume II (Advanced Techniques) (Department of
Geodetic Science, The Ohio State University, Columbus, Ohio).
Remondi, B.W. (1984). "Using the Global Positioning System *(GPS) Phase Observable
for Relative Geodesy: Modeling, Processing, and Results" Ph.D. dissertation available
as CSR-84-2 (Center for Space Research, The University of Texas at Austin, Austin,
Texas).
Robertson, D.S. (1975). Geodetic and Astrometric Measurements with Very-Long-Baseline
Interferometry X-922-77-228 (NASA GSFC, Greenbelt, MD).
Rotge, J.R., Shaw, G.L., and Emrick, H.W. (1985). "Measuring Earth Rate Perturbations
with a Large Passive Ring Laser Gyro" in Mueller, 1985, 2, 719-729.
Rykhlova, L.V. (1969). "Motion of the Earth's Pole, 1846.0-1891.5, from Observations at
Pulkovo, Greenwich, and Washington" Soviet Astronomy 12, no. 5, 898-899.
Sato, T. and Hanada, H. (1985). "A Program for the Computation of Ocean Tidal Loading
Effects 'GOTIC'" in Mueller, 1985 2, 742-748.
Scherneck, H.G. (1983). Crustal Loading Affecting VLBI Sites Department of Geodesy,
Report No. 20 (University of Uppsala, Institute of Geophysics, Uppsala, Sweden).
Schwiderski, E.W. (1978). Global Ocean Tides, Part 1, A Detailed Hydrodynamical Model
Report TR-3866 (U.S. Nav, Surface Weapons Center, Dahlgren, VA).
Schwiderski, E.W. (1980). "On Charting Global Ocean Tides" Rev. Geophys. Space Phys.
18, 243-268.
Schwiderski, E. (1983). "Atlas of Ocean Tidal Charts and Maps, Part I: The Semidiurnal
Principal Lunar Tide Af2" Marine Geodesy 6, 219-256. (See also Chapter 8.)
Shawe, M.E. and Adelman, A.G. (1985). "Precision Laser Tracking for Global and Polar
Motion" IEEE Trans, on Geosci. and Rem. Sens. GE-23, 4, 391-397.
4 / TERRESTRIAL COORDINATES AND THE ROTATION OF THE EARTH 277
Snyder, J.P- (1983). "Map Projections Used by the U.S. Geological Survey" Geological
Survey Bulletin 1532 2nd ed. (U.S. Government Printing Office, Washington, DC).
Soffel, M., Herold, H., Ruder, H., and Schneider, M. (1988). "Reference Frames in Rela-
tivists Space-Time" in Babcock and Wilkins, 1988, 99-103.
Sovers, O.J. and Fanselow, J.L. (1987). "Observation Model and Parameter Partials for the
JPL VLBI Parameter Software ' M ASTERFIT '- 1987" JPL Publication 83-39, Rev. 3.
Stephenson, F.R. and Lieske, J.H. (1988). "Changes in the Earth's Rate of Rotation
Between A.D. 1672 and 1806 as Deduced from Solar Eclipse Timings" Astronomy and
Astrophysics 200, 218-224.
Stephenson, F.R. and Morrison, L.V. (1984). "Long-term Changes in the Rotation of the
Earth: 700 B.C. to A.D. 1980" Phil. Trans. R. Soc. Lond., A 313, 47-70.
Teisseyre, R., ed. (1989). Gravity and Low-Frequency Geodynamics (Elsevier Science Pub-
lishers, Amsterdam.)
Thomas, P.D. (1952). "Conformal Projections in Geodesy and Cartography" U.S. Coast
and Geodetic Survey Special Publication No. 251 (U.S. Government Printing Office,
Washington, DC). Fifth printing, 1968.
Torge, W. (1980). Geodesy, C. Jekeli, Trans. (Walter de Gruyter, New York, NY).
Vondrak, J. (1969). "A Contribution to the Problem of Smoothing Observational Data"
Bull. Astron, Inst. Czechoslovakia 20, 349-355.
Wahr, J.M. (1981). "The Forced Nutations of an Elliptical, Rotating, Elastic, and Ocean-
less Earth" Geophys. J. Roy. Astron. Soc. 64, 705-727.
Wahr, J.M. (1985). "Deformation Induced by Polar Motion" J. Geophys. Res. 90, 9363-
9368.
Wahr, J. (1987). "The Earth's C21 and S21 gravity coefficients and the rotation of the
core" Geophys. J.R. Astr. Soc. 88, 265-276.
Wahr, J. (1990). Correction to the above paper, Geophys. J. Int. In press.
Yoder, OF., Williams, J.G., and Parke, M.E. (1981). "Tidal Variations of Earth Rotation"
J. Geophys. Res. 86, 881-891.
Yumi, S. and Yokoyama, K. (1980). "Results of the International Latitude Service in a
Homogeneous System" (International Polar Motion Service, Mizusawa, Japan).
Zelensky, N., Ray, J., and Liebrecht, P. (1990). "Error Analysis for Earth Orientation
Recovery from GPS Data" in McCarthy et al. (1990).
CHAPTER
Orbital Ephemerides
of the Sun, Moon, and Planets
by E.M. Standish, XX Newhall, J.G. Williams, and D.K. Yeomans
5.1 FUNDAMENTAL EPHEMERIDES
The fundamental ephemerides are the basic computation, based on the equations
of motion and fit to the observational data, of the positions and velocities of so-
lar system bodies. Fundamental ephemerides are the bases for computing apparent
ephemerides, representational ephemerides, phenomena, orbital elements, and sta-
bility characteristics. The fundamental planetary and lunar ephemerides of The
Astronomical Almanac are DE200/LE200. constructed at the Jet Propulsion Labo-
ratory in 1980. They result from a least-squares fit to a variety of observational data
followed by a numerical integration of the dynamical equations of motion governing
the major bodies of the solar system. DE200/LE200 are the bases for computing
the apparent ephemerides and phenomena that are listed in the almanac.
This chapter describes the equations of motion used for the major bodies in the
solar system, the numerical integration program used to compute the motions, the
observatiotial data to which the ephemerides were adjusted, the reference frame in
which the ephemerides were adjusted, the reference frame in which the coordinates
are expressed, and the estimated accuracies of the resultant positions.
Numerical integration of the equations of motion is currently the most accurate
method of computing fundamental ephemerides. The accuracy is limited by the
accuracy of the observational data, by the completeness of the model of the solar
system used, and, if integrating for a very long time period, by the round-off and
truncation errors of the numerical integrator.
Although analytical or numerical theories can be developed as fundamental
ePhemerides, the number of terms required for current observational accuracy is
Prohibitive. Rather, such theories are currently used for limited-accuracy ephemer-
>des for the sake of convenience.
279
280
EXPLANATORY SUPPLEMENT
Representative epheoierides to specified accuracies can be developed based on
Chebyshev polynomials alone and in combination with reference orbits. Mean orbit
elements can be used to compute approximate positions and velocities.
Historical information about former methods of computing ephemerides is also
provided.
5.11 Gravitational Model
The gravitational model used in DE200 includes all the known relevant forces acting
upon and within the solar system. A relevant force is defined as any force that
produces an observable or measurable effect.
Included in the equations of motion were the following: (1) point-mass in-
teractions among the Moon, planets, and Sun; (2) general relativity (isotropic,
parametrized post- Newtonian), complete to order 1 / c2; (3) Newtonian perturba-
tions of a few selected asteroids; (4) Moon and Sun action upon the figure of the
Earth; (5) Earth and Sun action upon the figure of the Moon; (6) Earth tide action
upon the Moon; and (7) physical libration of the Moon. The equations for these
effects are described in detail in Section 5.21. The relativistic effects on the time
arguments are discussed in Chapter 2.
5.12 The Ephemeris Reference Frame
The positions and velocities of the ephemerides are referenced with respect to iner-
tial space, since the equations of motion are defined with respect to inert ial space.
Any systematic error present in the ephemeris positions or motions should be re-
garded as ephemeris error, not as a rotation of the coordinate reference frame.
The orientation of the reference frame of DE200/LE200 is described by Standish
(1982). Since the dynamical equinox is a quantity inherent in the ephemerides, it
is possible to derive its location from the ephemeris itself. The alignment of the
ephemeris reference frame with the J2000 equinox of the ephemerides then becomes
relatively straightforward. Thus, the dynamical equinox, inherent in an ephemeris,
is not equivalent to the equinox of any specific stellar or radio-source catalog.
5.13 The Astronomical Constants Used in the Ephemerides
Associated with the ephemerides is the set of astronomical constants used in the
creation of the ephemerides. Many of these values do not agree exactly with the
standard IAU 1976 set of constants. The difference has been necessary in order to
provide a best fit of the ephemerides to the observational data. These constants are
associated directly with the ephemerides and are considered to be an integral part
of them. They are presented in Section 5.42.
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 281
5.2 COMPUTATION OF EPHEMERIDES
5 21 Mathematical Model
Newhall et al. (1983) present the equations of motion integrated for the creation
of the planetary and lunar ephemerides, DE200/LE2OO. That description is repro-
duced here. The mathematical model includes contributions from: (a) point-mass
interactions among the Moon, planets, and Sun; (b) general relativity (isotropic,
parametrized post-Newtonian); (c) Newtonian perturbations of a few selected aster-
oids; (d) Moon and Sun action on the figure of the Earth; (e) Earth and Sun action
on the figure of the Moon; (f) Earth tide action on the Moon; and (g) physical
libration of the Moon.
5.211 Point-Mass Interactions The principal gravitational force on the nine plan-
ets, the Sun, and the Moon is modeled by considering those bodies to be point
masses in the isotropic, parametrized post-Newtonian (PPN) n-body metric (Will,
1974). Also included are Newtonian gravitational perturbations from the asteroids
Ceres, Pallas, Vesta, Iris, and Bamberga. These five asteroids were found to have
the most pronounced effect on the Earth-Mars range over the time span covered by
the Mariner 9 and Viking spacecraft range measurements. The n-body equations
were derived from the variation of a time-independent Lagrangian action integral
formulated in a nonrotating solar-system barycentric cartesian coordinate frame.
The reference plane is the Earth's mean equator of B1950.0. The jc-axis coincides
approximately with the 1950 equinox of the FK4 catalog.
For each body i, the point-mass acceleration is given by Equation 5.211-1
f _ y Mrj ~ r-) f . 2(0 + 7) W (20 ~ 1) Hk (Vi\2
'point mass ,3, 1 1 c2 2^ r c2 Z-j fik 1\c)
/va2 2(1+7) 3 Hn - tj) ■ tj]2 l , . .. \
jjti 'J
3 + 47 \ - fyfj \ Mrm -r.)
2c2
^2^111 ^3 — , (5.211-1)
FV m=i r'"'
where r,-, ih and f, are the solar-system barycentric position, velocity, and acceler-
ation vectors of body i; fij = Gnij, where G is the gravitational constant and trij is
the mass of body j; r$ = \rj — r;|; @ is the PPN parameter measuring the nonlinear-
ity in superposition of gravity; 7 is the PPN parameter measuring space curvature
Produced by unit rest mass (in this integration, as in general relativity, (3 = 7 = 1)'
v'-' = |r,|; and c is the velocity of light.
282
EXPLANATORY SUPPLEMENT
In the last term on the right-hand side of Equation 5.211-1, quantities em-
ploying the index m refer to the asteroids. The positions of the asteroids are not
integrated but are obtained from polynomials representing heliocentric Keplerian
ellipses. The polynomials give good representations for perturbations on the planets
at the present time. At times in the distant past the polynomials drifted from the
real asteroid orbits, but the perturbations were smaller (Williams, 1984) for Mars
than any ancient optical-measurement accuracy.
The quantity r,- appearing in two terms on the right-hand side of Equation 5.211-1
denotes the barycentric acceleration of each body j due to Newtonian effects of the
remaining bodies and the asteroids.
5.212 Figure Effects Long-term accuracy of the integrated lunar orbit requires
the inclusion of the figures of the Earth and Moon in the mathematical model. In
DE200 the gravitational effects due to figures include:
(1) The force of attraction between the zonal harmonics (through fourth degree)
of the Earth and the point-mass Moon and Sun;
(2) The force of attraction between the zonal harmonics (through fourth degree)
and the second- and third-degree tesseral harmonics of the Moon and the
point-mass Earth and Sun.
The mutual interaction between the figures of the Earth and Moon is ignored.
The contribution to the inertial acceleration of an extended body arising from
the interaction of its own figure with an external point mass is expressed in the
^rjC, coordinate system, where the f-axis is directed outward from the extended
body to the point mass; the 7j-axis is directed east (lying in the selenographic xy-
plane, perpendicular to the f-axis); and the £-axis is directed north, completing the
right-hand system (see Figure 5.212.1). In that system (Moyer, 1971),
T
( "1
V
If
J.
♦EOT
(n+ l)P„(sin<£) "
0 (5.212-1)
- cos 4>P',,(sin 4>)_
—{n+ \)F"(s\n 4>)[Cnm cos mX + Snm sin mX]
m sec <AP"(sin 0)[-Cnm sin mX + S,m cos mX]
=i |_ cos 4>P","(sin 4>)[C„,„ cos mX + Snm sin mX]
where n is the gravitational constant G times the mass of the point body; r is the
center-of-mass separation between the two bodies; n\, and n2 are the maximum
degrees of the zonal and tesseral expansions, respectively; P„{sm <j>) is the Legendre
polynomial of degree «; P"'(sin 4>) is the associated Legendre function of degree n and
order m\ J„ is the zonal harmonic for the extended body; Cnm, Snm are the tesseral
harmonics for the extended body; a is the equatorial radius of the extended body;
(p is the latitude of the point mass relative to the body-fixed coordinate system in
which the harmonics are expressed; and A is the east longitude of the point, mass
in the same body-fixed coordinate system.
The primes denote differentiation with respect to the argument sin <p. The ac-
celerations are transformed into the solar-system barycentric cartesian system by
application of appropriate rotation matrices: first by a rotation from the j*^ system
to the selenographic system, followed by the application of the inverse libration-
angle matrix.
The interaction between the figure of an extended body and a point mass also
induces an inertial acceleration of the point mass. If f fig denotes the acceleration
given in Equation 5.212-1 when expressed in solar-system barycentric coordinates,
then the corresponding acceleration Tpm °f the point mass is
rPM = -^rFIG, (5.212-2)
PPM
where ^FtG and /iPM are the gravitation constant G times the masses of the extended
body and point mass, respectively.
5.213 Earth Tides The tides raised by the Moon on the Earth appear as a bulge
leading the Earth-Moon line by a phase angle 6. The resulting geocentric acceler-
ation of the Moon is given by the expansion
3k2tim ( . Vm\ ( a
Ttides ? 1 + ~
' em
r,,
x + y6
y — xb
z
(5.213-1)
284 EXPLANATORY SUPPLEMENT
where fcj is the potential Love number of the Earth; ae the radius of the Earth;
rem is the geocentric lunar distance; and x, y, and z are the geocentric cartesian
coordinates of the Moon expressed in the true-of-date system. (The rotation to
mean equator and equinox of J2000.0 is performed by the application of an inverse
nutation matrix followed by an inverse precession matrix. The nutation matrix is
evaluated using only the leading (18.6-yr) term.) In addition, (j,m is the gravitation
constant times the mass of the Moon; fie is the gravitation constant times the mass
of the Earth.
The inertial accelerations follow from the conservation of the center of mass.
For further discussion, see Williams et al. (1978).
5.214 Lunar Librations It is necessary to form a matrix transforming between
the inertial coordinate system of the integration and the selenographic system. The
Euler angle definitions and their differential equations were taken from Goldstein
(1950). <f> is the angle along the Earth's fixed equator from the fixed equinox to
the line of nodes with the Moon's true equator; 9 is the inclination of the Moon's
true equator to the Earth's fixed equator; and tp is the angle along the Moon's
equator from the line of nodes to the reference meridian of the selenographic system.
Following customary procedures, we define
a C-A
= (5.214-1)
and
B-A
1L = (5.214-2)
where A, B, and C are the three principal moments of inertia of the Moon and
C > B > A. The relationship between 0L, -yL, J2, C22 and C / ma2 (where m is the
mass of the Moon and a is the lunar radius) is described by Ferrari et al. (1980).
Let FpxG be the force on the Moon due to the gravitational interaction of
the lunar figure and an external point-mass Earth or Sun. Ffig is derived from
Equation 5.212-1 . Then the torque N on the Moon is given by
N = r x Ffig, (5.214-3)
where r is the vector from the lunar center of mass to the point mass. In the
selenographic principal- axis system the equations for the angular velocity vector 0
are related to the Euler angles through
iox = tf> sin 9 sin ip + 9 cos tp
uiy = <p sin 9 cos tp — 9 sin tp (5.214-4)
ujz- 4> cos 9 + 4>.
ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 285
Table 5.214.1
Lunar Libration Angles and Rates
Epoch Epoch
JD 2440400.5 JD 2444400.5
tp 5.13326253997944214 -003 -3.88899107819826403 -002
| 1 . 005339296430386 1 8 - 004 - 1 , 27558 1 40671 24741 3 - 004
9 3. 82365872861 401 043 - 001 4. 3091 2863340948250 - 001
0 1 . 441 67407096036579 - 005 5. 09287249525085733 - 005
ip 1.29422680846346332 + 000 1.33180551146558290 + 000
ip -1.19916273482678420 - 004 9.33300094985815842 -005
The differential equations for the angular velocity come from Euler's equations.
wvw, + —
1 - PlIL A
Cjy = /Jib** + ^ (5.214-5)
j.N*
g
c
Finally, the differential equations for the three Euler angles are
dix sin tp + iMy cos ip + 9(tp - 4> cos 9)
5n1
9 = 6jx cos ib - uy sin ip - <pip sin 9 (5.214-6)
$ = qz - ij) cos 9 + (p9 sin 9.
The second-order differential equations for the Euler angles of a rigid body Moon
are integrated numerically, considering torques induced by the Earth and Sun.
The libration states at the epoch 2440400.5 are given in Table 5.214.1. From
the nature of its definition the angle $ (denoting the angular displacement of the
selenographic meridian from the Earth's equator) grows rapidly with time, changing
by 2t radians per lunar revolution. To limit the magnitude of the numbers carried
by the numerical integrator, a linear polynomial was removed from the initial con-
ditions for ip presented to the integrator. In all calculations the actual values of ib
and tp used are:
(5.214-7)
286 EXPLANATORY SUPPLEMENT
where il'-mt(t), i/'w(f) denote the angle and rate provided by the integrator:
0, =0.22997 1502 1898 189 rad d"1. (5.214-8)
Discussions of numerically integrated physical librations are given by Williams
et al. (1973), Cappallo (1980), and Cappallo et al. (1981). Nonrigid-body effects
are described by Yoder (1979) and Cappallo (1980). The numerically integrated
physical librations used to fit the lunar laser data came from the program described
by Williams et al. (1973) rather than from that described here, since higher accuracy
is needed for the data reduction than for the integration of the lunar orbit.
5.215 Solar-System Barycenter In the n-body metric, all dynamical quantit ies are
expressed with respect to a center of mass whose definition is modified from the
usual Newtonian formulation. The solar-system barycenter is given by Estabrook
(1971, private communication) as
= 0, (5.215-1)
where
lc2 ' 2c2 ^r«\
In Equation 5.215-2 fij is defined as before and v, is the barycentric speed of body
nj = \rj - n\. (5.215-3)
During the process of numerical integration only the equations of motion for the
Moon and planets were actually evaluated and integrated. The barycentric position
and velocity of the Sun were obtained from Equation 5.215-1. It should be noted
that each of Equations 5.215-1 and 5.215 -2 depends on the other, requiring an
iteration during the evaluation of the solar position and velocity.
5.22 Numerical Integration
The numerical integration of Equations 5.211-1 and 5.214-6 was carried out using
a variable-step-size, variable-order Adams method (Krogh, 1972). The maximum
allowable order of any of the 33 equations is 14; the actual order at any instant is
determined by a specified error bound and by the behavior of backward differences
of accelerations.
5 / ORBITAL EPHEMERIDES OF THE SUN. MOON. AND PLANETS 287
5 221 Force Model Evaluation The calculation and arrangement of accelerations
at each integration step is as follows:
(1) The integrator subroutine provides new states (positions and velocities) for
the nine planets, the Moon, and the libration angles.
(2) The asteroid states are evaluated from fixed polynomials.
(3) Equations for the relativistic masses are evaluated for the planets, Moon,
asteroids, and Sun, using current states for all bodies except the Sun. (The
barycentric state of the Sun calculated at the end of the previous step is
retained for this evaluation.)
(4) The present approximate state of the Sun is obtained from the constraint
(Equation 5.215-1).
(5) Equations 5.215-2 are evaluated again, using this new estimate of the solar
state.
(6) Equation 5.215-1 is evaluated a second time to provide the current state of
the Sun.
(7) Equations 5.211-1 are evaluated to obtain the accelerations of the nine
planets and the Moon.
(8) It has proved numerically more suitable to integrate the lunar ephemeris
relative to the Earth rather than to the solar-system barycenter. The solar-
system barycentric Earth and Moon states are replaced by the quantities
rem and rfi, given by
rm = Tm~te (5.221-1)
and
rB = ^rc + M'"r'", (5.221-2)
He + (*m
where the subscripts e and m denote the Earth and Moon, respectively.
Note that rem is the difference of solar-system barycentric vectors and is
distinguished from a geocentric vector by the relativistic transformation
from the barycenter to geocenter. (The vector can be interpreted as
representing the coordinates of the Newtonian Earth-Moon barycenter rel-
ative to the solar-system barycenter. It has no physical significance and
does not appear in force calculations; it is solely a vehicle for improving
the numerical behavior of the differential equations.)
(9) The equations for the libration angle accelerations are evaluated.
5.222 Estimated Integration Error The method of error control used in the in-
tegration puts a limit on the absolute value of the estimated error in velocity of
each equation at the end of every integration step. Step size and integration orders
were adjusted on the basis of estimated error. The limits selected for DE200 were
2 X 10" 17 AUd-1 in each component of the equations of motion for the planets and
Moon, ancj 2 x 10-15 rad d_1 for each component of the libration equations.
288 EXPLANATORY SUPPLEMENT
5.223 Adopted Constants The integration requires inputting the numerical val-
ues of a number of parameters. Some of these parameters, such as the initial po-
sitions and velocities of the planets and Moon, result from the least-squares fits
and are different in each fit. Other parameters, such as some of the masses and the
Earth's zonal harmonics, come from outside sources and are only rarely changed
for these present purposes. Some parameters, such as the mass of the Earth-Moon
system, can be derived from the data, but for convenience are changed only when
statistically significant improvements can be made over the standard values.
The lunar secular acceleration is an important parameter that has been subject
to uncertainty and confusion. Lunar laser data now provide an estimate of the lunar
secular acceleration (Calame and Mulholland, 1978; Williams et ai, 1978; Dickey
et at, 1982; and Dickey and Williams, 1982). Input can be a Love number and a
phase shift that are converted to an acceleration (Chapront-Touze and Chapront,
1988). The value used in DE118 (DE200) was h = -23?9 ± 1"3 / cy2. This may be
compared with the more recently estimated value of -24"9 ± I'.'O / cy2 (Newhall et
ai, 1989).
5.23 Orientation of Ephemerides
5.231 Procedure Summary The J2000-based ephemerides, DE200/LE200, were cre-
ated directly from the Bl950-based ephemerides, DE118/LE62, by a simple reori-
entation of the reference frame. The orientation is presented by Standish (1982a);
the essential features are reproduced here. This procedure was done in the following
five steps:
(1) The 1950 dynamical equinox of DEI 18 was determined to be
£n8(1950) = -ar(1950) = +0"5316, (5.231-1)
where the superscript signifies that the "rotating" definition of the equinox
(Standish, 1981) was used.
(2) DE118 was rotated onto its own dynamical equinox of 1950, producing
DE119:
rm = Rj(-0?5316)r]l8. (5.231-2)
(3) DEI 19 was precessed to the epoch J2000, using the 3 X 3 matrix P given
by Lieske (1979):
r'1]9=Pr119. (5.231-3)
(4) The dynamical equinox of DE119' at J2000 was determined to be E^,Q —
-0JQ0073.
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 289
(5) DEI 19' was adjusted onto its own dynamical equinox of J2000, thereby
producing DE200:
rm = Rz(0:!OO073)r,n9. (5.23 1-4)
5,232 Astronomical Constants The preceding analyses and comparisons reveal a
number of features that merit further discussion.
The origin of the reference system of DEI 18 should be approximately that of
the FK4, since the ephemeris has been fit to data that included transit observa-
tions of the U.S. Naval Observatory that have been referenced to the FK4. As such,
the determination of £1^(1950) may be interpreted as a determination of the FK4
equinox. The value of 0'.'5316 found in Step 1 of Section 5.231 agrees remarkably
closely with that of Fricke (1982) for the FK4: (EFJC4(1950) = 0'.'525). The smallness
of the difference of 0'.'0066 must be fortuitous, however, for the expected uncertain-
ties of the two determinations are nearly an order of magnitude greater.
One may calculate the mean obliquity, e, at a given epoch directly from the
ephemerides, using an analysis similar to the one used for computing Q. This deter-
mination should be quite accurate (± 0'.'01 ) since the data set used in the adjustment
of the ephemerides included 10 years of Lunar Laser-Ranging. As such, the obliquity,
especially at the mean epoch of the laser-ranging data (1975), is well represented
by the ephemerides. The analysis gives the result
400(/2000) = 23°26'21"4119, (5.232-1)
where the rotating (Standish, 1981) sense of the definition has been used.
This latter number is then to be compared with IAU (1976) value of
e,AU " 23°26'21,.,448, (5.232-2)
giving
£u,-^oo = 0"0361. (5.232-3)
Bretagnon and Chapront (1982) have made an independent analysis of DE200/LE200,
using their analytical planetary and lunar theories (Bretagnon, 1980, 1981; Chapront-
Touze, 1980; Chapront-Touze and Chapront, 1980). The analysis, covering 100
years, produces a value for the dynamical equinox (rotating sense) of 0','00068. This
independent check verifies that DE200/ LE200 is indeed referred to its own dynam-
ical equinox of J2000.0 to within an accuracy of ± OL'001 .
290 EXPLANATORY SUPPLEMENT
5.3 OBSERVATIONAL DATA FIT BY THE PLANETARY AND LUNAR
EPHEMERIDES
The planetary and lunar ephemerides represent the results of numerical integra-
tions of the equations of motion that describe the gravitational physics of the solar
system. It is assumed that the equations of motion accurately describe the physics,
at least to the presently observable accuracy. Also, the computer program that in-
tegrates these equations has been demonstrated to be sufficiently accurate. It is
the least-squares adjustment process, of course, that produces the initial conditions
and the set of astronomical constants that are used by the numerical integration
computer program.
The most critical feature of the construction of modern-day lunar and plan-
etary ephemerides is the accuracy and variety of the observational data to which
the ephemerides are fit. This section (taken from Standish, 1990) describes the
observational data used in the creation of DE200/LE200.
The observational data include meridian transits, satellite astrometry, radar
ranging to a planet's surface, ranging to various spacecraft, and laser ranging to
the lunar reflectors. The reduction formulas are given for each data type. The
accuracies of each type are given along with the numbers of observations and the
time spans covered by the different sets of data.
5,31 Optical Data
There have been three types of optical data used in the Jet Propulsion Laboratory
(JPL) ephemerides up to and including DE118. These are the transit observations
from the U.S. Naval Observatory (USNO), some astrometric plate data of Saturn
taken at the University of Virginia's Leander McCormick Observatory, and a set of
normal points of Neptune and Pluto provided by the USNO.
5.311 Transit Circle Data The only transit data used in the JPL ephemerides up
to and including DE118 were the observations from the U.S. Naval Observatory
(USNO) taken with the six-inch and nine-inch meridian circles. Furthermore, only
those since 1911 have been used, the date signifying the introduction of the imper-
sonal micrometer. (See Table 5.311.1.)
The observations are recorded in the publications of the USNO, second series.
The data listed in these publications have been reduced to the Washington catalog of
the concurrent epoch. Table 5.311.1 presents references for the Washington catalogs
in which the observations are given.
All the transit observations for catalogs W(25) through W4(50) were trans-
formed to the reference system of the FK4 using the formulas of Schwan (1977).
For catalog W(10) the table published in the second series of the U.S. Naval Ob-
servatory publications was used.
ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 291
Table 5.311.1
Transit Circle Observations from the U.S. Naval Observatory
that have been used in the JPL Ephemerides, DE118
Time-span
Telescope
Vol/Part
Number of
Observations
191 1-1Q1A
6"
XI
2436
1913-1925
9"
XIII
3381
W(25)
1925-1935
6"
XV l/l
6911
W(40)
1935-1944
9'
xv/v
4547
W(50)
1935-1941
6"
XVI/I
3777
W2(50)
1941-1 949
6"
XVI/III
3444
W3(50)
1949-1956
6"
XIX/I
3678
W4<50)
1956-1962
6"
XIX/II
4051
W5(50)
1963-1971
6"
XXIII 1 f 1
5811
(Cir.)
1975-1977
6"
1543
Totals
39579
Note: The Vol/Part refers to the Second Series of the USNO Publications
Data since the end of the W5(50) in 1971 have been provided in machine-
readable format by the USNO. These must be considered as provisional, since the
transit circle was refurbished during the interval 1972-1974.
5.3111 Basic reduction of the transit observations Transit observations are differ-
ential in nature — the planetary observations undergo the same processing as those
of the observed stars, both being related to the standard catalog of the epoch. The
observations are published as geocentric apparent right ascensions and declinations,
taken at the time of meridian passage. For comparison, then, one obtains a com-
puted position from the ephemerides by iterating to find the time at which the local
apparent hour angle of the planet is zero. The formulation for computing apparent
places has been essentially identical to that described in Chapter 3.
5.3112 Phase corrections A phase-effect correction is determined by the USNO
and applied to the observations before publication. This correction is based on the
theoretical effect that should be expected due to the geometric situation. Here,
empirical formulas have been used to fit and remove further phase effects from the
transit-circle observations. It seems that changes result mainly from "irradiation
effects" whereby an observer tends to measure a bright illuminated limb differently
from a darker terminator.
292 EXPLANATORY SUPPLEMENT
For the center of light measurements, for both Mercury and Venus, the USNO
correction was removed from the observations, and, instead, the following formula
was applied:
[Cq + CJ+CjF + CiI3], (5,3112-1)
Aa
s
sin 9
AS
r
cos 0
where / is the phase angle expressed in units of 90 degrees (/ = r/90); © is the position
angle of the midpoint of the illuminated edge; and i is the phase angle between the
Sun and the Earth, subtended at the observed body. The coefficients, Cq through
C3, are solved for in the ephemeris solutions. The resulting cubic polynomial agrees
quite well with that found by Lindegren (1977).
For an illuminated limb measurement, for both Mercury and Venus, the USNO
correction was retained, and the following formula was applied in addition:
Aa
sin©
AS
cos©
[ifl + Li/ + L2/2+L3/3].
(5.3112-2)
Here, the coefficients are in units of seconds of arc. For measurements of four limbs,
the USNO correction was retained and the following formula was applied in addi-
tion:
sin 0
AS
cos 0
Bk sin 2z,
(5.3112-3)
where k - 4, 5,..., 8 for Mars, Earth,..., Neptune. This empirical formula may be
compared with that of Chollet (1984), the form of which has been derived from
physical considerations.
The forms of these preceding "phase correction" formulas were all chosen
strictly from empirical considerations.
5.31 13 Day corrections For each Washington catalog, corrections are applied to
the observations obtained during daylight hours; namely, observations of the Sun,
Mercury, and Venus. Typically, these are given in tables as functions of the object's
declination and of the time of day. Day corrections have not been applied, however,
to the preliminary observations given in the USNO circulars. Therefore, for the
1975-1977 data the following day corrections were solved for and removed from the
observations of the Sun, Mercury, and Venus:
Aa =Ai + A2sin6 (5.3113-1)
AS = £>i + D2 sin 6, (5.3113-2)
where S is the declination of the body. A\, A2, D\, and Dj are empirical quantities.
6 / ORBITAL EPHEMER1DES OF THE SUN, MOON, AND PLANETS 293
5 3114 Catalog drift The mean motions of the inner planets are determined pri-
marily by the strength of the ranging data in the least-squares fit. Therefore, any
inconsistency between these mean motions and those determined from the optical
data will appear as secular trends in plots of the optical residuals. These drifts are
due partly to an incorrect value of precession and to an equinox motion in the FK4
reference system, which have been accounted for in two ways.
5.312 Corrections to precession and equinox drift The secular-like drifts in the
optical residuals have been modeled by the standard formulas,
Aa = {Ak + An sin a tan 6)750
AS = ( An cos a)T50, (5.312-1)
where T50 is the time in centuries past 1950. These corrections were subtracted from
the observed values of all the transit observations of the planets and the Sun.
If one assumes that these parameters come from precession error and equinox
drift exclusively, then the following relations apply:
Ak = —E + Ap cos e - Al and An = Ap sin e, (5.3 12-2)
where E is the equinox motion; Ap is the correction to lunisolar precession; Al is
the correction to planetary precession; and e is the value of the Earth's obliquity.
For DE118, the value of Art was set to +0"438/cy, corresponding to Ap s
+l"10/cy, Fricke's (1971) determination of the correction to precession. Ak was fixed
to -O'.'266/cy, corresponding to Fricke's (1982) determination of the FK4 equinox
motion, E - +l'.'275/cy. The value of AJfc should have been -O'.'237/cy, since the
correction for planetary precession of Al = — O'.'029/cy (see Lieske et at, 1977) was
inadvertently omitted.
With Ak and An constrained to the values determined by Fricke, the corrections
to the computed residuals of the transit data are identical to those that would
have been found if the whole procedure (observations and reductions) had been
performed in the J2000.0 (IAU 1976) reference system.
s-3121 Catalog offsets With At fixed for the solution to DE118, there was the
possibility of a residual secular trend in the optical residuals. It was decided to
introduce constant offsets in both right ascension and in declination for each of the
Washington catalogs. The values for these are included in Table 5.42.1.
s-313 Astrometric Plate Data for Saturn Over the years 1973-1979, the Univer-
sity of Virginia, under contract from JPL, provided measurements of photographic
Plates taken of the planet Saturn and its satellites (Ianna, 1974, 1980). The mea-
surements of the satellites and the background stars were provided to JPL and
294 EXPLANATORY SUPPLEMENT
were reduced by Standish, using the satellite ephemerides of Null (1978) and the
Zodiacal Catalog of the USNO (Schmidt et at, 1980). The "observed" positions
of the satellites from the reductions were then compared to "computed" positions
derived from the satellite and planetary ephemerides. Since the errors of the satel-
lite ephemerides were expected to be small in comparison to those of the planetary
ephemerides, and since the satellites errors enter in a quasi random way when many
observations are used, it was assumed that the differences (O — C$) could be at-
tributed to the errors in the planetary ephemerides. Accordingly, the differences
were used to adjust only the ephemeris of Saturn.
5.314 Normal Points for Neptune and Pluto As mentioned before, the main op-
tical data used by JPL are the transit observations from the USNO, beginning in
1911. Clearly, for the outermost planets this is not enough data to determine a full
orbit. Consequently the USNO transmitted a set of normal points that gave the
differences between one of their experimental ephemerides and an intermediary JPL
ephemeris (DEI 14). The optically based USNO ephemeris gave what was believed
to be reasonable fits to the observations of Neptune and Pluto. These data were
then used to adjust the solution for DE118. The data were transmitted in the form
of corrections to longitude and latitude, one pair of points every 400 days. The time
span for Neptune was 1846-1974; for Pluto, 1914-1974.
5.32 Radar-Ranging Data
5.321 Computation of Radar-Ranging Measurements Radar-ranging measurements
are actual timings of the round-trip light-time of the electromagnetic signal from
the time that it leaves the transmitter until the time that it arrives at the receiver
(Figure 5.321.1). The timing is done by an atomic clock at the observing station.
For an observation received at the time t (expressed in TDB time units), the
round-trip light-time is given by the difference UTC(/) - UTC(/ - Td - r„), where
t„ = |rR(r - rd) - rA(r - rd - tu)\ I c + Arjrel] + Ar„[cor] + AT„[tropo] (5.321-1)
and
rd = |rA(0 - rR(t - rd)\ I c + Ar^frel] + Ar^cor] + Ar</[tropo], (5.321-2)
tu and rd are the light-times (in TDB units) of the upleg and the downleg, respec-
tively; rA is the solar-system barycentric position of the antenna on the Earth's sur-
face; rR is solar-system barycentric position of the reflection point on the planet's
surface; c is the velocity of light; and the three Ar's are the corrections to the light-
times due to relativity, the electron content of the solar corona, and the Earth's
Figure 5.321.1
Diagram of geometry for planetary radar ranging
troposphere, respectively. The two formulas are solved iteratively, first for rd then
for tu.
The location of the antenna is computed in a straightforward manner using a
planetary ephemeris and the proper formulas of precession, nutation, tuning, and
polar motion with which one orients the Earth into the reference frame of the
ephemerides. The location of the point of reflection was computed by assuming the
surfaces of Mercury and Venus to be spherical. Radar ranging to Mars was not
included in the solution for DEI 18.
5.3211 Time delays for relativity, the solar corona, and the troposphere The time-
delay due to relativity, given by Shapiro (1964), is obtained by integrating along
the signal path over the value of the potential. For each leg of the signal path, the
delay is given by the formulas
Ar[rel] =
(1 + -))GM
In
e+p + q
e+p-q
(5.3211-1)
296
EXPLANATORY SUPPLEMENT
where 7 is the PPN parameter of general relativity and where e, p, and q are the
heliocentric distance of the Earth, the heliocentric distance of the planet, and the
geocentric distance of the planet, respectively. These distances are evaluated at
t — Td for the planet, at t - — r„ for the Earth during the upleg, and at ( for the
Earth during the downleg.
The delay from the solar corona (see Muhleman and Anderson, 1981) is ob-
tained by integrating along the signal path from point Pi to point P2 over the
density of ionized electrons, Ne (cm-3),
At [cor] =
(5.3211-2)
where c is the speed of light (cm/sec), / is the frequency (Hz), and s is the linear
distance (cm). The density is given by
r6 r2
where r is the heliocentric distance expressed in units of the solar radius and /? is the
solar latitude. The values for the constants, A, a, and b are included in Table 5.42.1.
The delay from the Earth's troposphere is discussed by Chao (1970). For each
leg, it is
Ar[tropo] = 7 nsec / (cos z + 0. 0014 / (0. 045 + coU)), (3.3211-4)
where z is the zenith distance at the antenna.
S.322 Radar-Ranging Data The strength of a radar echo from a planet's surface
varies as the inverse fourth power of the distance. Up to the present time, high-
quality planetary radar-ranging observations have been confined to times when the
distances are less than about 1 AU. At these distances, the precision of the ranging
observations has been seen to be about 100 meters. However, for ephemeris mea-
surements, variations in the topography of the planet's surface introduce variations
into the observations which tend to dominate the uncertainties. For Mercury and
Venus these amount to about 1 km.
The radar-ranging measurements used in the JPL ephemerides have come
from five antennas. These are located in Arecibo, Puerto Rico; Tyngsboro, MA
(Haystack); Westford, MA (Millstone); and Goldstone, CA (DSS 13 and DSS 14).
In addition, some of the measurements at Goldstone were taken in the bistatic
mode: transmitting with DSS 14 and receiving with DSS 13. Table 5.322.1 presents
a list of the measurements which have been used directly in DEI 18.
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 297
Table 5.322.1
Radar-Ranging Observations Used Directly in DE118
Planet Timespan Antenna Number
of Observations
Mercury
1966-1971
Arecibo
106
1966-1971
Haystack
217
1971
Gold 13/14
9
1972-1974
Gold 14
30
Mercury Totals
362
Venus
1966-1970
Arecibo
246
1966-1971
Haystack
219
1964-1967
Millstone
101
1964-1970
Gold 13
294
1970-1971
Gold 13/14
14
1971-1973
Gold 14
44
1973-1977
Gold 14
25
Venus Totals
945
5.33 Spacecraft Range Points
There were three sets of spacecraft-ranging data used in the ephemeris solutions
for DE118. They are (1) normal points from the Pioneer Missions to Jupiter, (2)
normal points to the Mariner 9 orbiter of Mars, and (3) actual range measurements
to the Viking Landers on the surface of Mars.
Normal points represent modified distance measurements. The original round-
trip range and Doppler measurements have been reduced using the JPL Orbit
Determination Program (Moyer, 1971). This reduction is an adjustment for various
parameters, including the spacecraft orbit, the planet's mass and gravity field, and
so forth. As such, the resultant range residuals represent derived corrections to the
nominal planetary ephemeris used in the reduction. These residuals are then added
to the geometric (instantaneous Earth-planet) range in order to give a pseudo
"observed" range point.
5-33l Pioneer Normal Range Points to Jupiter There was one normal range point
from each of the two Pioneer encounters of Jupiter. The major uncertainty for each
Point comes from the uncertainties in the determination of the spacecraft's orbit
298
EXPLANATORY SUPPLEMENT
Table 5.332.1
Mariner 9 Orbiter Normal Range Points to Mars
Dates
a
Number of
(usee)
Observations
2441272-2441361
0.25
77
2441389-2441540
0.29
81
2441541-2441555
0.76
487
2441577-2441602
Total
645
Note: The four sets of points are grouped according to the
proximity in time to the solar conjunction of Mars (JD 2441568).
with respect to the center of mass of the planet. The observations, transmitted by
Null (1976), are as follows:
Pioneer 10: 1973 Dec 04 0" ET (2442020.50) / 2754747323+.40 /xsec.
Pioneer 11: 1974 Dec 03 0h ET (2442384.50) / 2439811990+10 ^sec.
These values represent one-way geometric (instantaneous) distances from the center
of mass of the Earth to the barycenter of the Jovian planetary system.
5.332 Mariner 9 Normal Range Points to Mars During its lifetime, the onboard
range transponder of the Mariner 9 Orbiter allowed accurate range measurements
to the planet Mars. These data exist in three sets according to their proximity to
the Martian solar conjunction (JD 2441568) when the 2300-MHz ranging signal
passed within 4 solar radii of the Sun at the heliographic latitude of +79°. When
near conjunction the major uncertainties of these points are due to uncertainties in
the densities of the ionized electrons in the solar corona through which the signal
passed; away from conjunction, the orbital uncertainties dominate. As shown in
Table 5.332.1, the three sets of data are weighed accordingly. The data have been
corrected for the solar corona using the formula given in Section 5.32.
5.333 Viking Lander Range Data The most accurate of all planetary position data
are the two-way ranging measurements taken of the Viking Landers on the surface of
Mars. Though only a single frequency (2300 MHz} was used, the solar corona could
be calibrated by using the nearly simultaneous dual-frequency measurements of the
Viking Orbiters. Observations were made on the average of once per week, typically
with about six range points per day. The residuals are seen to have a scatter of only
2 to 3 meters about the mean of the day; the means for each day, however, show
a scatter of about 6 meters among themselves. Without orbital uncertainties being
present, the dominating contributions to these residuals come from three sources:
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 299
the calibration of the solar corona, the calibration of the inherent time-delays in
the tracking station antennas (done before and after each pass), and the calibration
of the transponders of the landers themselves (done before launch).
The range data were reduced using the formulas given in Section 5.321 with
the reflection point on the planet's surface being represented by the location of
the lander on Mars. For this, one needs the Martian coordinates of the landers as
well as a set of angles used to express the orientation of Mars within the ephemeris
reference frame. The position of the lander, expressed in the frame of the ephemeris,
is given by
r = R.t(-s)R;(-I?)R,(-/)RJ(-J?;)R,(-/;)R;(-V')r0, (5.333-1)
where e is the obliquity of the ecliptic; Q and / are the longitude and latitude of the
mean Martian orbit upon the ecliptic; J?9 and Iq are the mean node and inclination
of the Martian equator upon the mean orbit; V is the longitude of the Martian
prime meridian measured along the equator from the intersection of the orbit; and
ro, the Mars-fixed coordinates of the lander, and where
Q'q = flq-M\ = + Ac and V' = V + Aip cos l'q, (5.333-2)
where Ait' and A= express the nutation of Mars, computed from the formulation
of Lyttleton et al. (1979). The Mars-fixed coordinates of the lander are computed
from the cylindrical coordinates,
Tq _ [w cos A, u sin A, v]T. (5.333-3)
A summary for the Viking Lander ranging data is as follows:
Lander #1 2442980-2444054 Points: 683
Lander #2 2443026-2443417 Points: 78.
The values for the parameters used in the reductions are included Table 5.42.1.
The values for e, J?, and / were adopted: those for Qq, Iq, and V, as well as the
coordinates of the landers, were estimated in the least-squares adjustments.
5'34 Lunar Laser Range Data
The Lunar Laser Range (LLR) data consist of time-of-flight measurements from
McDonald Observatory to any one of four retroreflectors on the Moon and back
again. The retroreflectors are at the Apollo 11, 14, and 15 landing sites and on
300 EXPLANATORY SUPPLEMENT
the Lunakhod 2 vehicle. These 2954 range points are distributed from August 1969
to June 1980. The normal points through 1973 have been published by Abbot et
al. (1973), Mulholland et al. (1975), and Shelus et al. (1975). The LLR data are
deposited in the National Space Science Data Center. During the least-squares fit
the ranges have been weighted according to the instrumental errors that accompany
each point. The general trend is toward improving accuracy with time. The simple
post-fit rms residual is 31 cm in one-way range. The weighted rms residual is 27cm
(Dickey et al, 1982).
For the Moon, the mean orbital elements that affect its geocentric distance
will be strongly determined by the laser-ranging data. The eccentricity is implicitly
determined to a few parts in 109 and the mean anomaly to a few milliarcseconds.
There is a strong determination of differential geocentric lunar and solar ecliptic
longitudes resulting from strong solar perturbations on the lunar distance {ampli-
tudes of about 3000 km and 4000 km for the two leading terms). During the span of
the observations the differential longitudes are known to a few milliarcseconds. Out-
side the data span the error in the lunar longitude is dominated by the uncertainty
in the lunar tidal acceleration. (See Williams and Standish, 1989.)
5.4 LEAST-SQUARES ADJUSTMENT OF THE EPHEMERIDES
The least-squares adjustment for DE118/LE62 involved 175 explicit unknown pa-
rameters and 50424 observational equations. The adjustment was done in two steps.
Using a previous ephemeris (DE111/LE55) as the nominal ephemeris, a preliminary
solution was made involving all solution parameters that were judged to be relevant.
Then, the values for certain parameters were rounded off and forced into the final
solution for DE118/LE62 which readjusted, slightly, the remaining parameters.
There are a number of reasons for forcing certain parameters to have specific
values: the adoption of someone else's, presumably more accurate, determination of
the constant (e.g., the mass of Pluto); the rounding of values for cosmetic reasons
(e.g., a truncated value for the inverse mass of Jupiter); or the need for consistency
with outside sources (e.g., the 1AU values of Ait and An).
The creation of DE118/LE62 was done partially with the intent that this
ephemeris (when rotated onto the J2000 equator and equinox) would be used for the
foundation of future national almanacs. Ideally, the ephemeris should have incorpo-
rated all of the astronomical constants of the IAU (1976) J2000 reference system.
However, it was recognized that more modern values for some constants were nec-
essary in order to fit the more accurate observational data properly. In particular
it was not possible to use the IAU set of constants and still produce acceptable
solutions for the Viking Lander ranging data or the Lunar laser-ranging data.
ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 301
Table 5.41.1
The Observational Data Used for the Adjustment of DE1 18/LE62
1 y pu ui vuwii »fl uui i
I rtt Aran 1
II ILGI VOI
NumKpr nf
Oh qp Pint i ct
Standard D^vintinn
Wasninyion udnoiifa
1 Q1 1— 1 Q77
i y i i — i y / /
ouo / y
1 "fl f'Run Mar \/a.n\
n ^ I'M a re Mnni
U.D ^iviars, ...j iNSpj
c nil irn netrnnnfltni
oHturn asiromeuy
1 Q7Q—1 Q7Q
i y / o — i j / j
**■/ yu
U* o
Neptune, Pluto normal points
1825-1974
386
0".5
Radar-ranging
1964-1977
1307
1 0 km (before 1 967)
1 .5 km (after 1 967)
Pioneer ranges at Jupiter
1973-1974
2
3km, 12km
Mariner 9 ranges at Mars
1971-1972
645
40-1 00 m
Viking Lander ranges
1976-1980
761
7m
Lunar laser ranges
1969-1980
2954
18cm
Total
50424
5.41 The Observational Equations
Table 5.41.1 gives the number of observational equations for each different set of
data. Each equation was normalized by multiplying it by the factor, 1 / (Tq, where
ctq is the a priori standard deviation of a single observation of a particular set of
data. Previous experience in working with the various sets has led to a knowledge
of the individual accuracies of each type of observation. The values of the a priori
standard deviations, listed in Table 5.41.1, were chosen to be approximately equal
to the rms post- fit residuals.
5.42 The Solution Parameters
Table 5.42.1 gives the values and the formal standard deviations of the solution
parameters of the least-squares adjustment for DE118/LE118. The parameters for
which values were adopted from outside sources are enclosed in square brackets.
Parameters that were determined from the preliminary solution, rounded, and then
forced into the final solution are enclosed in parentheses. The adjustment of the or-
bital parameters is inherent in the values of the initial conditions of the ephemerides
that are listed in Table 5.42.2.
In addition to the parameters listed in the table, there were a number of pa-
rameters pertaining to the plate solutions and catalog adjustment of the Saturn
astrometrical data. These entered into the solutions implicitly; they were present in
the solution just as if they had been carried along explicitly; however, their presence
is not seen directly.
Table 5,42.1
Values and Formal Standard Deviations of the Constants
Used in DE118/LE118 (DE200/LE200)
Cralo P^iyfnr Hrm/AI tl
OCdlB PdClUr [Hm/AUJ
\ 1 tiJOl?/ O / U.DC
± 0.002)
Earth/Moon Mass Ratio
[81.300587]
Sun/Plane1 Mass Ratios
Sun/Mercury
= [6023600]
Sun/Venus
= [408523. 5]
Sun/EM-harv
= [328900. 55]
Sun/Mars
= [3098710]
Sun/Jupiter
= (1047. 35010.0001
Sun/Saturn
= (3498.0 ±0.014)
Sun/Uranus
= (22960 ± 6)
Sun/Neptune
= [19314]
Sun/Pluto
= [130000000]
Asteroid GM value
GM(Ceres)
= [0.1746x 10"
1a]
[Air/day^]
GM(Pallas)
= [0.3200x10"
13i
GM(Vesta)
= [0.4080x 10"
13,
GM(lris)
= [0.160Qx10~
14,
GM(Bamberga) = [0.2600x 10
14i
Radii
Mercury =
2439. 990 ± 0. 093
Venus =
6051.813 ± 0. 058
Mars =
[3397.515]
Day Corrections
*1 =
0:'268 ± 0'.,048
A2 =
01378 ±01160
Pi =
01153 + 01048
Ps = -
11131 ± 0" 1 60
Phase Corrections
c0 = -
■0.03 + 0.017
Ci =
0.39 ± 0. 087
c2 =
0.47 ±6.11
c3 = -
-0.26 ± 0.039
kl = "
0'.'15 + 0'.'067
Li =
2181 ± 0:'30
L2 = -
2145 ± 0:'37
1-3 =
0:73 ± on 3
B4 =
o:'40 ± 01022
B5 =
Oi'97 + 0l'077
B6 =
o:74 ± o:'14
B7 =
0'.'20 ± 0'.'30
B8 =
0'.'93 ± 0'.'49
Catalog Drift
Ak = I
-07266 /cy]
An = [0:'438/cyJ
Catalog Offsets
W10
0'.'365
01303
W20
-ovoog
-01033
W2S
oiosi
01084
W40
-01027
-01015
W150
-01028
01090
W250
-01102
01094
W350
-o:ois
-0'.'021
W450
-01106
-0'.'001
W550
-01087
-0'.'032
WCIR
-01231
-0'.'020
ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS
303
Table 5.42.1, continued
Values and Formal Standard Deviations of the Constants Used in
DE1 1 8/ LE11 8 (DE200/LE200)
Solar Corona [ cm 3]
A
1.22+ 03 x 10s
a
0.44±.02x106
b
0.44±.03x106
Mars Orientation
V
[deg]
[328. 70742325]
V
[deg/day] 350. 891 99047 + 0. 00000024
!q
[deg]
25. 1808415 ± 0.0013
'!«
[deg/cy]
0. 030 ± 0. 006
nq
35. 3371555 + 0. 0026
fin
4
[deg/cy]
-0.118 + 0. 010
1
[deg]
[1.85]
)
[deg/cy]
[-0. 00820]
fl
[deg]
(49. 17193]
fl
[deg/cy]
[-0. 29470]
E
[deg]
[23. 445789]
e
[deg/cy]
[-0.01301]
Viking Lander
"1
[km]
3136.515 10.001
«*1
[km]
1284.587 ±0.024
*1
(deg]
311.8027 ± 0. 0033
uz
[km]
2277.374 ±0. 002
[km]
2500.1 84 ±0.022
A2
[deg]
1 34.0343 + 0. 0034
Note: Numbers in square brackets were not solved for but adopted from other
sources. Numbers in parentheses are rounded values determined from a
preliminary solution and then forced unchanged into the present final solution.
5.43 The Standard Deviations
The standard deviations listed in Table 5.42.1 are the formal values, straight from
the least-squares solutions. They do not represent realistic uncertainties, for it is
well known that formal uncertainties tend to be overly optimistic, sometimes by an
order of magnitude. This is a direct result of incorrect or incomplete modeling, either
through the equations of motion or through the data reductions. In the present case,
some of the sources of the incompleteness are known. For discussions of what are
believed to be realistic uncertainties associated with the planetary ephemerides, see
Newhall et al. (1983), Standish (1985), and Williams and Standish (1989).
5-5 NUMERICAL REPRESENTATION OF THE EPHEMERIDES
The ephemerides produced by the Jet Propulsion Laboratory are obtained by nu-
merical integration of the equations of motion of the Moon and planets. The imme-
diate output from the integrating program is not suited to subsequent rapid eval-
uation of ephemeris quantities. To provide ephemerides convenient for distribution
Table 5.42,2
The Initial Conditions of the Ephemerides at JED 2440400.5 in au and Au/day
Coordinates
Velocities
J. 0/<iDUil £o4b9b3715— 001
k\
3. 36784520455775328
-003
Yi
-9. 15490552856159762-002
fa
2 4889342837585R4ft0
— on?
*1
-8.59810041345356578-002
Z\
1.29440715971588809
-002
\ f& ni ic-
venus
x2
b. Ua*;494j77bo441072— 001
*2
1 . 095241 993547441 85
-002
y2
-3. 49132444047697970-001
¥2.
1 5612506911 "547704?
002
- 1 . 955443255802 1 7404 -00 1
22
6. 32887643692262960
-003
cann— Moon
XB
^ i Cm ACt H ~7f\A A C A A~7EQ
1 . 1 bUI 491 7044544758—001
*B
1 . 681 1 6200395885947
-002
barycenter
Yb
-9. 26605558053098135-001
Vo
1 74ini?fi1fi1fiQ4eiQQ
Zb
-4. 018062651 17824489-001
7.55975079765192612
-004
Msrs
x$
— i. 14boaab54b*;040833— 001
1 . 44820048365775564
-002
Va.
-1 . 32836653338579221 +000
jA
2 372fl'i174,5fifl7?nic;i
— uut
-6. 06155187469280320-001
24
-2. 83748756861 61 1 822
-004
Jupiter
*5
—5. 38420864140637830+000
*5
1 . 09236745067075960 -
-003
y5
-8. 31 249997353602621-001
JO
—6 52329390316976699
-003
*>
-2. 25098029260032085-001
-2.82301211072311896-
-003
Saturn
f, Oo9oo94<:o73227478+000
-3.21 7205 1 41 22007756 -
-003
ye
4, 59570992672261 122+000
Vb
4 33063208949070? 1fi-
-003
2&
1 . 55842916634453457+000
26
1,92641681926973271-
-003
1 1 ri 1 1 1>
*7
— 1 . 8<io969 1 1 379855 1 69+00 1
*7
2. 21 544461 295879596 -
-004
>7
— 1. 16273304991353263+000
if-*
31
—3 76765491 Bfi3fi47T51
^7
-2. 50376504345852463-001
'2l
- 1 . 65324389089726956 -
-003
Mrint'i infl
1 N i - |Jlunc
*B
— 1 . b059504334 1 7291 60+00 1
2. 64312595263412502-
-003
ys
-2. 39429413060150989+001
Vc
/8
— 1 5034fi6fiR4'iR4fi?n71 -
-003
3>
-9. 4004277295751 4666+000
-6. 81268556592018307-
-004
Pli ttn
Xg
—3. 04879969725404637+001
Xg
3. 22541 768798400992-
-004
-8, 7321 6536230233241 -001
/9
—3 14875QQ6£i,i41Q?fl7A-
•003
*9
8. 91 135208725031935+000
*9
-1.08018551253387161-
003
Ql in
4. 50479585567460 182-003
*S
-3. 52445744568339381 -
007
ys
7. 73254474689074171 -004
ys
5. 17763778067222128-
006
■5s
2. 68503998557327098-004
*s
2.22911325240040085-
006
Moon
-8. 08177235835125058-004
6.01084831482911873-
004
yM
-1, 99463003744199594-003
yM
-1 . 6744546915006061 9-
004
A*
- 1 . 087262721 62086794-003
-8. 5562081 0990486240-
005
Note: Coordinates given are heliocentric coordinates for plenets, geocentric coordinates for the Moon, and
solar-system barycentric coordinates for the Sun.
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON. AND PLANETS 3Q5
and use throughout the astronomical community, an efficient procedure was de-
veloped for the generation of coefficients representing an interpolating polynomial.
This section (taken from Newhall, 1989) describes the generation and accuracy of
those coefficients.
The numerical integration process carries the states of the Sun, Moon, and
planets as their solar-system barycentric Cartesian coordinates in the inertial J2000
reference frame. The integration method used is a variable-step-size, variable-order
Adams procedure.
5.51 Cbebyshev Polynomials
Chebyshev polynomials are the functions of choice for ephemeris interpolation. They
are stable during evaluation, they give a near-minimax representation, and they
provide a readily apparent estimate of the effect of neglected terms on interpolation
error (For a thorough treatment of these polynomials see Rivlin, 1974).
The nth Chebyshev polynomial T„(t) is denned by the recursion formula
Tn(t) = 2tTn^(t) - r„_2(/), »*2,3, ... (5.51-1)
with To(t) = 1 and T\ (r) = /. The applicable range of t for interpolation is — 1 < / < 1 .
Any function f(t) has an approximate Mh-degree expansion in Chebyshev poly-
nomials
N
fU) = J2a"T^' (5.51-2)
and, when differentiated,
N
/(/) = £>„?;(/), (5.51-3)
where the a„ are chosen in a manner appropriate for /(/) and /(/). In the present
case, where f(tj) and f(tj) denote a coordinate and its derivative computed at dis-
crete times tj by the integrating program, the a„ serve to define the function /(/)
as a polynomial. The task becomes the determination of a set of a„ that pro-
vide interpolated values suitably approximating those available from the original
backward-difference representations carried by the integrator. The following section
details the generation of the a„ for ephemeris body coordinates.
5.52 Chebyshev Coefficient Generation
An algorithm was developed that creates Chebyshev polynomial coefficients repre-
senting the Cartesian coordinates of the ephemeris bodies. The full span of, say,
306 EXPLANATORY SUPPLEMENT
60 years of an epheineris file is segmented into contiguous intervals, or granules, of
fixed length. (The actual length of a granule depends on the body; see Table 5.53.1
for details.) For each coordinate of an ephemeris body the algorithm produces the
Chebyshev coefficients an that define the interpolating polynomial valid over a given
granule. There are as many sets of coefficients generated for each coordinate as there
are granules covering the ephemeris span.
fn the procedure, nine pairs of position and velocity values are obtained for
each granule at equally spaced times — one pair at each end point and seven in the
interior. The output is the set of a„ for the polynomial that is an exact fit to the
end points of the set of positions and a least-squares fit to the interior positions.
The differentiated polynomial is an exact fit to the end points of the velocity set
and a least-squares fit to the interior velocities. (The derivation of this procedure
is given in Newhall, 1989.)
This approach has the advantage that interpolated position and velocity are
continuous at the common end point of adjacent granules; it also minimizes the
effect of noise that would otherwise degrade the interpolated velocity obtained from
differentiation of a polynomial based on position values alone.
5.53 Interpolation Error and Polynomial Degree
It is essential to have a quantitative estimate of the maximum error expected from
the interpolation process when the polynomials described previously are used to
extract coordinate values at arbitrary times. (It should be noted that the term
"error" here refers to the difference between interpolated and integrator-supplied
values; it does not indicate the degree of accuracy to which the original integrated
ephemeris represents the dynamical state of the solar system.)
The Chebyshev polynomials provide a convenient and reliable estimate of in-
terpolation error. An arbitrary function has the exact representation as an infinite
Chebyshev expansion
f{t) = YjanTn(t). (5.53-1)
The maximum value of T„(t) is unity on the interval [-1, 1], the domain of validity
for interpolation. Therefore, when a function is approximated by an A/th-degree
polynomial, as in Equation 5.51-2, the maximum error s arising from the omitted
remainder of the series has the upper bound
< ^2 M\TM $ £ W- (5.53-2)
ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 307
Table 5.53.1
Granule Length and Polynomial Degree for the 1 1 Ephemeris Bodies
Granule
Polynomial
Body
length (days)
degree (A/)
t\A&rr \ i rv
ivici uui y
8
13
Vsnus
1 6
g
CdllM — IVIUUII U d 1 y L. t?l lit? 1
16
12
Mars
I u
Jupiter
32
7
Saturn
32
6
Uranus
32
5
Neptune
32
5
Pluto
32
5
Moon
4
12
Sun
16
10
Investigation has shown that the granule length and the polynomial degree N can
be chosen so that \a„+i I a„\ = 0. 1 or less for n > N. This implies that the maxi-
mum expected interpolation error is about one tenth the magnitude of the highest
retained coefficient a„.
The accuracy criterion for standard JPL ephemerides is that the interpola-
tion error for all coordinate values must be less than 0.5 millimeters. (The DE102
ephemeris covers 4400 years; in the interests of providing a significantly compressed
file, the interpolation-error criterion was relaxed.) The minimum degree N of the
interpolating polynomial is 3, as the requirement, that the end-point position and
velocity values be matched exactly yields four constraints; the 18 combined position
and velocity values permit a maximum degree of 17. Table 5.53.1 lists the granule
length and polynomial degree for each body on the .IPL ephemeris files.
5.6 COMPUTATION OF OBSERVATIONAL EPHEMERIDES
Computing the fundamental ephemerides as described in Sections 5.2 to 5.4 pro-
duces ephemerides in a space-fixed coordinate system at a given epoch in terms of
the barycenter of the solar system and in barycentric time. From these ephemer-
ides, observational ephemerides can be computed that are called apparent geocen-
tric ephemerides, with respect to a fixed epoch and a reference star catalog. Transit
times of the objects and astronomical phenomena can be computed as well.
308
5.61 Apparent Positions
EXPLANATORY SUPPLEMENT
The apparent position of a planet or the Moon represents the position as it would
be seen from the center of mass of the Earth at some date in the coordinate sys-
tem defined by the Earth's true equator and equinox of date, if the Earth and its
atmosphere were massless, transparent, and nonrefr acting. The apparent position
could then be related to observables for a particular instrument at some location on
the surface of the Earth that is rotating and contains a refracting atmosphere. In
addition to the transformation of the coordinate systems, the time argument must
be changed from bary centric time to Terrestrial Time. The formula and methods
for performing the transformation from barycenter to geocenter are provided in
Section 3.32. Reduction to topocentric coordinates, with the effects of refraction
included, is given in Section 3.7. The relationships between barycentric time and
Terrestrial Time are given in Section 2.22.
5.62 Astrometric Positions
Astrometric positions are useful for faint objects such as Pluto, minor planets, and
comets where observations are made directly with respect to the star positions on
some fixed epoch in the same field. Thus, if a photographic plate or a CCD image is
taken with respect to star positions that are known on a catalog for a fixed epoch,
it is desirable to relate the solar-system object to those positions. An astrometric
position is therefore easier to compute than an apparent place since the differential
position is not concerned with the orientation of the coordinate system. The details
concerning computation of astrometric positions are given in Section 3.62.
5.63 Transit Ephemerides
For observational planning and for certain types of observations, it is desirable
to know when the body will be on the meridian. Because Terrestrial Dynamical
Time (TDT) is independent of the rotation of the Earth and because the rotation
of the Earth is not precisely predictable, it is not possible to calculate or publish
in advance the times of transits over a fixed geographical meridian. Therefore, an
auxiliary reference has been introduced. The position of the ephemeris meridian in
space is conceived as being where the Greenwich meridian would be, if the Earth
rotated uniformly with the rate implicit in the definition of Terrestrial Dynamical
Time. Thus, the ephemeris meridian is 1.002738 AT east of the actual meridian of
Greenwich, where AT is the difference TDT - UT. When referred to the ephemeris
meridian, phenomena depending on the rotation of the Earth may be calculated in
terms of TDT by the same methods as those by which calculations referred to the
Greenwich meridian are made in terms of Universal Time.
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 309
Ephemeris transit, is the time when the right ascension of the object is equal to
the value determined by substituting the TDT time into the equation for sidereal
time. The times of transit are expressed in TDT and refer to the transits over the
ephemeris meridian. Because the longitude of the ephemeris meridian (expressed in
time measure) is 1.002738 AT* east of the Greenwich meridian, the TDT of transit
over the Greenwich meridian is later than that of ephemeris transit by
AT/ (1 - 0.99727 X rate of change of R.A.) (5.63-1)
or approximately,
ATX 1 + rate of change of R. A . ) (5.63-2)
if the right ascension is expressed in the same unit as the time interval. The UT
of Greenwich transit is therefore later than the tabulated TDT of the ephemeris
transit by the small quantity:
AT x rate of change of R.A., (5.63-3)
which may, at present, reach about 2s for the Moon, but is less than 0!3 for the
planets.
For the Sun the TDT of ephemeris transit is simply 12n minus the equation
of time interpolated to the time of transit; the UT of Greenwich transit differs by
0. 002738A7.
For the Moon, the TDT of upper or lower transit of the Moon over a local
meridian may be obtained by interpolation of the tabulated time of upper or lower
transit over the ephemeris meridian, where the first differences are about 25 hours.
The interpolation factor p is given by
p = A + 1. 002738 AT, (5.63-4)
where A is the east longitude and the right-hand side is expressed in days. (Divide
longitude in degrees by 360 and AT in seconds by 86400). In general, second-order
differences are sufficient to give times to a few seconds, but higher order differences
must be taken into account if a precision of better than 1 second is required. The
UT of transit is obtained by subtracting AT from the TDT of transit, which is
obtained by interpolation.
For planets the UT of transit over a local meridian may be obtained from:
time of ephemeris transit - (A / 24) x first difference.
(5.63-5)
31 0 EXPLANATORY SUPPLEMENT
This produces an error that is usually less than 1 second, where A is the east
longitude in hours and the first difference between transit times is about 24 hours.
5.7 ORBIT AND EPHEMERIDES OF OTHER BODIES
WITHIN THE SOLAR SYSTEM
5.71 Minor Planets and Comets
Orbital information is available for several thousand minor planets (often called
asteroids) and several hundred cornets. Orbital data for all numbered minor planets,
and ephemerides for those objects that reach opposition during the volume year, are
published annually by the Institute for Theoretical Astronomy in St. Petersburg,
Russia. These annual volumes, entitled Ephemerides of the Minor Planets, can be
purchased from the White Nights Trading Company, 520 N.E. 83rd St., Seattle,
WA 98115, USA.
The Minor Planet Center (60 Garden St., Cambridge, MA 02138, USA) offers
a variety of services for obtaining information on minor planets and comets. Cata-
logues of Cornetary Orbits are published every few years. Approximately ten times
each year, astrometric observations, ephemerides, and updated orbits are published
for both comets and minor planets in the Minor Planet Circulars. These circulars
are available in printed form, by electronic mail, and as machine-readable files on
personal computer disks. Magnetic tapes, containing large files of astrometric data
for thousands of minor planets and comets, can also be purchased from The Minor
Planet Center.
The accuracy of published orbital parameters for minor planets and comets
varies from object to object. In general, objects whose orbits are based upon op-
tical observational data spread over lengthy time intervals will have more orbital
accuracy than those objects whose orbits are based upon shorter data intervals.
The inclusion of radar data in the orbit determination process will dramatically
improve the orbital accuracy, particularly if the existing optica! data are spread
over only a short interval of time.
The annual Astronomical Almanac volumes publish ephemerides for the first
four numbered minor planets. These volumes also include orbital elements, mag-
nitude parameters, opposition dates, and opposition apparent magnitudes for the
larger minor planets that reach opposition during the volume year. In addition, the
osculating orbital elements are given for the periodic comets that are scheduled to
pass perihelion within the volume year.
The orbital parameters given for the minor planets and comets can be used to
generate approximate ephemeris information using two-body formulations. These
same elements can be employed to generate more accurate ephemerides when used
to initialize numerical integrations that include planetary perturbations. For active
5 / ORBITAL EPHEMERIDES OF THE SUN. MOON. AND PLANETS 31 1
short-period comets, the generation of accurate ephemerides often requires that,
jn addition to the planetary perturbations, the comet's rocketlike outgassing forces
be taken into account. The mathematical models used to describe these so-called
nongravitational effects are described by Marsden et al, (1973) and Yeomans and
Chodas (1989).
The magnitude parameters (H, G) given for the minor planets can be used to
compute the object's apparent magnitude using the following relationships:
apparent magnitude -H+5 log(Z>) - 2. 5 log[(l - G)P, + GP2\ (5.71-1)
Pi = exp[-3.33(tana/ 2)0 63]
P2 = exp[-1.87(tana/2)L22], (5.71-2)
where
D = geocentric distance in AU
r = heliocentric distance in AU
a = Sun-asteroid-Earth angle.
In the Astronomical Almanac, the ephemeris positions and orbital elements
are given with respect to a coordinate system defined by the equinox of the FK5
fundamental reference star catalog and the J2000.0 epoch (FK5/J2000.0 system),
whereas the vast majority of solutions for orbital elements, prior to 1992, were
computed with respect to the equinox of the FK4 reference star catalog and the
B1950.0 epoch (FK4/B1950.0 system). In Sections 5.711-5.714, the expressions
"1950.0" and "2000.0" refer to the FK4/B1950.0 and FK5/J2000.0 systems respec-
tively. Comission 20 of the International Astronomical Union has recommended the
following conversion procedures between the two systems.
5.711 Conversion of Comet and Asteroid Astrometric Positions from B1950.0 to
J2000.0 Let an = object's right ascension referred to B1950.0 system.
do = object's declination referred to B1950.0 system.
a = object's right ascension referred to J2000.0 system.
d = object's declination referred to J2000.0 system.
1. Calculate the rectangular components of the object's position vector Tq.
cos(«o) cos(<fo)
sin(fl0)cos(rf0) (5.711-1)
sin(rf0)
12
EXPLANATORY SUPPLEMENT
2. Remove the effects of elliptical aberration (E-terms) to form astrographic
position vector (r\).
Let Tq be the vector transpose of Tq. Then form Fj .
T[ = r0 - A + (ro ■ A)r0 where A =
-1.62557 x 10-61
-0. 31919 x 10-6
-0. 13843 x 1CT6
(5.711-2)
and (r<j • A) is a scalar product.
3. Form the J2000.0 vector r from the vector rj using the matrix M;
r = Mr,,
(5.711-3)
where
M = X(0) + 7*X(0);
T = Julian centuries from B1950.0 epoch to the observation time (r)
= (/- 2433282. 423)/ 36525.
The equation r = Mri is a modification to Equation 30 in Murray (1989),
where the following matrices are those defined as his equations 28 and 29.'
X(0) =
10bX(0) :
0. 9999256794956877 -0.011181 4832204662
0. 0 1 1 1 8 1 483239 1717 0. 9999374848933 1 35
0. 0048590037723 1 43 - 0. 000027 1 702937440
-0. 0026455262 — 1. 1539918689 2. 1 1 1 1 346 1 90
1.1540628161 -0.0129042997 0.0236021478
-2.1112979048 -0.0056024448 0.0102587734
-0.0048590038153592"
-0.0000271625947142
0.9999881946023742
(5.711-4}
The matrix expression (rX(0)) removes the error introduced when the
B1950.0 reference stars were advanced to the epoch of observation using
their proper motions (i.e., the error results from the incorrect constant of
precession and the equinox drift between the FK4 and FK5 systems). The
matrix X(0) corrects for the half arcsecond equinox offset of the FK4 at
the B1950.0 epoch and precesses the vector r{ from the epoch of B1950.0
to that of J2000.0 using the 1976 IAU constant of precession.
Calculate the new J2000.0 right ascension and declination (a,d) from r(*,y,z)
using the following expressions.
Let r = tJx2 +y2 + z2
cos(a) cos(d) =
sin(a) cos(rf) =
sin(af) =
(5.711-5)
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 313
5 712 Conversion of Comet and Asteroid Astrometric Positions from J 2000.0 to
B1950.0 Let <3o = object's right ascension referred to J2000.0 system.
do - object's declination referred to J2000.0 system.
a = object's right ascension referred to B 1950.0 system.
d = object's declination referred to B1950.0 system.
1. Calculate the rectangular components of the object's position vector r0,
(5.712-1)
r0 =
cos(ao) cos(^o)
sin(ao)cos(^o)
sin(</0)
2. Calculate the 3 x 1 vector *\{x\,y\,z\).
t\ - M_1ro where ro is a 3 x 1 vector. (5.712-2)
The M matrix is given in Section 5.711.
3. Include the effects of elliptic aberration (E-terms).
r = r i + A - (t\ ■ A)r, , (5.712-3)
where the vector A was defined above in Section 5.711 and (t/-A) is a scalar
product.
4. Calculate the new right ascension and declinations (a, d) using the following
expressions.
Let r = tJx2 +y2 +z2
r = t(x,y,z)
x
cos(a) cos(d) = -
r
sin(a) cos(d) = -
I r
sin(J)=-. (5.712-4)
r
5-713 Conversion of Inertia! Equatorial Coordinates from B1950.0 to J2000.0 Let
r2, v2 = equatorial rectangular position and velocity coordinates in J2000.0 system.
ri> Vj = equatorial rectangular position and velocity coordinates in B1950.0 system.
r2 = X(0)r1;
V2 = X(0)v!, (5.713-1)
Where X(0) is the 3x3 matrix defined in Section 5.711.
The inverse transformation from J2000.0 to B1950.0 is performed using the
'nverse (or transpose) of the X(0) matrix.
314
EXPLANATORY SUPPLEMENT
5.714 J2000.0 Keplerian Orbital Elements from B1950.0 Elements Given the six
Keplerian orbital elements a, e, T0, W, S2, and t, in the B1950.0 reference system,
one may compute the corresponding orbital elements for the J2000.0-based system
(denoted by primes):
a, a
e, e
To, n
si, q
i, t
= Semi-major axes in B1950.0, J2000.0 system
as Eccentricity in B1950.0, J2000.0 system
= Perihelion passage time in B1950.0, J2000.0 system
= Argument of perihelion in B1950.0, J2000.0 system
= Longitude of the ascending node in B1950.0, J2000.0 system
= Inclination in B1950.0, J2000.0 system.
Then a' = a, e — e', Tq - - Tq and u>', SI1, and f are computed from the following
expressions:
sin(u/ — u>) sin(/') - sin(7) sin(L + Si)
cos(u/ - ui) sin(/') = sifl(i) cos(7) + cos(/) sin(V) cos(L + SI)
cos{/') = cos(7) cos(/) - waif) sin(J) eos(L + SI)
sin(L' + SI') sin(/') = sin(/) sin(L + Q)
cos(L' + J?') sin(/') = cos(7) sin(7) + sin(/) cos(y) cos(L + SI),
(5.714-1)
Figure 5.714.1
The relationship between the B 1950.0 and J2000.0
reference frames and the orbitaf plane
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS
where the known angles, L', L, and J, are given by
315
Ll = 4?50001688
L = 5? 19856209
7 = 0f00651966.
To compute the corresponding angles w, SI, and / (B1950.0) given the angles w',
J?', and /' (J2000.0), use the following expressions;
sin(n/ — us) sin(7) = sin(7) sin(i' + Q')
cos(w' - uj) sin(/) = sin(/') cos(7) - cos(/') sin(7) cos(L' + Q')
cos(/) = cos(Z') cos(7) + sin(/') sin(7) cos(L' + Si') (5.714-2)
sin(L + Q) sin(/) = sin(/') sin(L' + S2')
cos(L + SI) sin(/) = - cos(/') sin(7) + sin(/') cos(7) cos(l' + S2').
The values for the constants U, L, and J have been determined using the fol-
lowing values for the obliquity of the ecliptic;
23?44578787 for (B1950.0)
23?43929111 for (J2000.0)
5.8 KEPLERIAN ELEMENTS FOR THE POSITIONS OF THE MAJOR
PLANETS
Lower accuracy formulas for planetary positions have a number of important appli-
cations when one doesn't need the full accuracy of an integrated ephemeris. They
are often used in observation scheduling, telescope pointing, and prediction of cer-
tain phenomena as well as in the planning and design of spacecraft missions. Also,
they are more easily incorporated into a computer program.
Classical Keplerian orbital elements are given in Table 5,8.1. The approximate
maximum errors over the period 1800-2050 are shown in Table 5.8.2. The errors
made when extrapolating outside this interval can be substantially greater. The
errors of the approximate formulas may be compared to those of the integrated
ephemerid es, DE200, which are less than 1 arcsecond throughout 1800-2050, often
much smaller.
Table 5.8.1
Classical Keplerian elements at the epoch J2000 (JED 2451545.0), given with respect to
the mean ecliptic and equinox of J2000. The angles are in degrees, and their centennial
rates are in arcseconds per century. The six elements are the semi-major axis,
eccentricity, inclination, longitude of ascending node, longitude of perihelion, and mean
longitude, respectively
a e t .0 & L
Mercury
0.38709893 0.20563069 7.00487 48.33167 77.45645 252.25084
0.00000066 0.00002527 -23.51 -446.30 573.57 261628.29 +415 rev
Venus
0.72333199 0.00677323 3.39471 76.68069 131.53298 181.97973
0.00000092 -0.00004938 -2.86 -996.89 -108.80 712136.06 +162rev
E-M barycenter
1.00000011 0.01671022 0.00005 -11.26064 102.94719 100.46435
-0.00000005 -0.00003804 -46.94 -18228.25 1198.28 1293740.63 +99 rev
Mars
1.52366231 0.09341233 1.85061 49.57854 336.04084 355.45332
-0.00007221 0.00011902 -25.47 -1020.19 1560.78 217103.78 +53rev
Jupiter
5.20336301 0.04839266 1.30530 100.55615 14.75385 34.40438
0.00060737 -0.00012880 -4.15 1217.17 839.93 557078.35 +8 rev
Saturn
9.53707032 0,05415060 2,48446 113.71504 92.43194 49.94432
-0.00301530 -0.00036762 6.11 -1591.05 -1948.89 513052.95 +3 rev
Uranus
19.19126393 0.04716771 0.76986 74.22988 170.96424 313.23218
0.00152025 -0.00019150 -2.09 1681.40 1312.56 246547. 79 + 1 rev
Neptune
30.06896348 0.00858587 1.76917 131.72169 44.97135 304.88003
-0.00125196 0.00002514 -3.64 -151.25 -844.43 786449.21
Pluto
39.48168677 0.24880766 17.14175 110.30347 224.06676 238.92881
-0.00076912 0.00006465 11.07 -37.33 -132.25 522747.90
Table 5.8.2
Approximate maximum errors of the Keplerian formulas over the
interval 1800-2050, given in heliocentric right ascension,
declination, and distance.
Right Ascension Declination Distance
["]
[1000 km]
["]
[1 000 km]
["]
[1000 km]
Mercury
20
6
5
1
5
1
Venus
20
10
5
2
20
10
Earth
20
15
5
4
20
15
Mars
25
25
30
30
40
40
Jupiter
300
1000
100
350
200
600
Saturn
600
4000
200
1400
600
4000
Uranus
60
800
25
400
125
2000
Neptune
40
800
20
400
100
2000
Pluto
40
1100
10
250
70
2000
s / ORBITAL EPHEMERIDES OF THE SUN. MOON, AND PLANETS 317
s.9 BASIS FOR PRE-1 984 EPHEMERIDES
prior to 1984, the ephemerides for the Sun, Mercury, Venus, and Mars were based
on the theories and tables of Newcomb (1898). Computerized evaluations of the
tables were used from 1960 through 1980. From 1981 to 1983, the ephemerides
•were based on the evaluations of the theories themselves. The ephemerides of the
Sun were derived from the algorithm given by S, Newcomb in Tables of the Sun
(Newcomb, 1898). Newcomb's theories of the inner planets (1895-1898) served as
the basis for the heliocentric ephemerides of Mercury, Venus, and Mars. In the case
of Mars, the corrections derived by F.E. Ross (1917) were applied.
Ephemerides of the outer planets, Jupiter, Saturn, Uranus, Neptune, and Pluto,
were computed from the heliocentric rectangular coordinates obtained by numeri-
cal integration (Eckert et a/., 1951). Although perturbations by the inner planets
(Clemence, 1954) were included in the printed geocentric ephemerides of the outer
planets, they were omitted from the printed heliocentric ephemerides and orbital
elements.
The lunar ephemeris, designated by the serial number j = 2, was calculated
directly from E.W. Brown's algorithm instead of from his Tables of Motion of the
Moon (1919). To obtain a strictly gravitational ephemeris expressed in the measure
of time defined by Newcomb's Tables of the Sun, the fundamental arguments of
Brown's tables were amended by removing the empirical term and by applying to
the Moon's mean longitude the correction
- 8'.'72 - 26!'74 T - 1132 T2, (5.9-1)
where 7" Is measured in Julian centuries from 1900 January 0.5 ET = JED 2415020.0.
In addition, this ephemeris was based on the IAU (1964) System of Astronomical
Constants, and was further improved in its precision by transformation corrections
(Eckert et al, 1966, 1969). The expressions for the mean longitude of the Moon and
its perigee were adjusted to remove the implicit partial correction for aberration
(Clemence et al, 1952).
5-91 Introduction of New Constants
In 1964 the International Astronomical Union adopted the 1968 IAU System of
Astronomical Constants (IAU Trans., 1966). The complete list of these constants,
with detailed explanation, reference and formulas were published in the Supplement
to the Astronomical Ephemerides 1968, which was reprinted in the 1974 edition of
the Explanatory Supplement.
By 1970 it was recognized that the ephemerides being used for the national
almanac publications required improvements — that the fundamental catalog of star
Positions (FK4) should be replaced by a new fundamental catalog of star positions,
318 EXPLANATORY SUPPLEMENT
that better values were known for some astronomical constants, that the definition
and practical realization of ephemeris time was inadequate and inconsistent, that
atomic time was available, and that a new epoch might be introduced to replace
B1950.0.
Since the necessary changes would affect many aspects of astronomy, it was
decided to introduce as many changes as possible at one time in a consistent system,
rather than to introduce them at many different times with resulting inconsistencies.
Agreement was reached that all changes should be introduced into the national
publications of the ephemerides for the 1984 editions, and that the changes should
be effective on 1984 January 1.
The changes were designed to be the best that can be achieved on the basis
of current knowledge and should reduce the systematic errors between theory and
observations by several orders of magnitude. However, it was recognized that per-
fection cannot be achieved and the improvements introduced should permit the
determination of further improvements, after the systematic deviations are reduced
by the present improvements.
The new resolutions form a self-consistent and interrelated system of changes
necessary to improve the astronomical reference system. The resolutions should
not be introduced individually or selectively. They must be implemented simulta-
neously. It is intended that whenever the new standard epoch ( J2000.0) is used, all
the computations and reductions will be based on the new astronomical constants
and resolutions. Furthermore, proper utilization of such data requires adherence to
the resolutions in any calculations or procedures.
The majority of the resolutions were prepared and adopted by the General As-
sembly of the IAU at the 1976 and 1979 meetings {IAU Trans., 1977, 1980). The
1979 IAU Theory of Nutation was adopted at the 1979 IAU General Assembly to
provide an improved nutation theory for astronomical computations. The IUGG
requested that the IAU reconsider its recommendations, not because of the inade-
quacy of the nutation theory, but because the Earth model underlying the nutation
theory was not considered an adequate model. Although differences in the Earth's
models could not be demonstrated to significantly affect the nutation, the relevant
commissions of the IAU have adopted the 1980 IAU Theory of Nutation. The Final
Report of the Working Group on Nutation was published by Seidelmann (1982).
The origin of the FK5 with respect to the FK4 was specified by Fricke (1982).
The new definition of the formula relating Greenwich mean sidereal time to Uni-
versal Time was published by Aoki et al. (1982).
The formulation using the new constant of precession was published by Lieske
et al. (1977) and Lieske (1979).
The 1976 IAU System of Astronomical Constants is given in Chapter 15. The
constants used in the creation of the planetary and lunar ephemerides, DE200/LE200,
are also given in Chapter 15. These are not all identical to those of the 1976 IAU
System, for it had become apparent in fitting the ephemerides to the observational
g / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 319
data, that it was necessary to adjust some of the constants in order to produce a
best-fit in the least-squares sense.
510 REFERENCES
Abbot, ILL, Shelus, P.J., Mulholland, J.D., and Silberberg, E.G. (1973). "Laser Observa-
tions of the Moon: Identification and Construction of Normal Points for 1969-1971"
Astron. J. 78, 784-793.
Adams, A.N., Bestul, S.M., and Scott, D.K. (1964). "Results of the Six-Inch Transit Circle,
1949-1956" Publ. U.S. Naval Obs. XIX, I.
Adams, A.N. and Scott, D.K. (1968). "Results of Observations Made with the Six-Inch
Transit Circle, 1956-1962" Publ U.S. Naval Obs. XIX, II.
Aoki, S., Guinot, B., Kaplan, G.H., Kinoshita, H.H., McCarthy, D.D., and Seidelmann,
P.K. (1982). "The New Definition of Universal Time" Astron. Astrophys. 105, 359-
361.
Bretagnon, P. (1980). "Theorie au deuxieme ordre des planetes inferieures" Astron. As-
trophys. 84, 329-341.
Bretagnon, P. (1981). "Construction d'une theorie des grosses planetes par une methode
iterative" Astron. Astrophys. 101, 342-349.
Bretagnon, P. and Chapront, J. (1981). "Note sur les formules pour le calcul de la
precession" Astron. Astrophys. 103, 103-107.
Bretagnon, P. (1982). "Theorie du mouvement de I' ensemble des planetes. Solution
VSOP82" Astron. Astrophys. 114, 278-288.
Bretagnon, P. and Chapront, J. (1982). Private communication in Standish (1982a).
Bretagnon, P. and Francou, G. (1988). "Planetary Theories in Rectangular and Spherical
Variables VSOP 87 Solution" Astron. Astrophys. 202, 309-315.
Brown, E.W. (1919). Tables of the Motion of the Moon (Yale University Press. New Haven,
CT).
Calame, O. and Mulholland, J.D. (1978a). "Lunar Tidal Acceleration Determined from
Laser Range Measurements" Science 199, 977-978.
Calame, O. and Mulholland, J.D. (1978b). in Tidal Friction and the Earth's Rotation P.
Broche and J. Sunderman, eds. (Springer, New York) p. 43.
Cappallo, R.J. (1980). "The Rotation of the Moon" Ph.D. Thesis. (MIT, Cambridge).
Cappallo, R.J., King, R.W., Couselman, C.C. Ill, and Shapiro, LI. (1981). "Numerical
Model of the Moon's Rotation" Moon and Planets 24, 281-289.
Chao, C.C. (1970). "A Preliminary Estimation of Tropospheric Influence on the Range
and the Range Rate Data During the Closet Approach of the MM71 Mars Mission"
Jet Prop. Lab Tech. Memo #391-129 (JPL, Pasadena, CA).
Chapront, J. (1970). "Construction d'une theorie litterale planetaire jusqu'au second ordre
des masses" Astron. Astrophys. 7, 175-203.
Chapront, J. and Simon, J.L. (1972). "Variations seculaires au premier ordre des elements
des quatre grosses planetes. Comparaison avec Le Verrier et Gaillot" Astron. Astro-
phys. 19, 231-234.
Chapront-Touze, M. (1980). "La solution ELP du probteme central de la Lune" Astron.
Astrophys. 83, 86-94.
Chapront-Touze, M. and Chapront, J. (1980). "Les perturbations planetaires de la Lune"
Astron. Astrophys. 91, 233-246.
320
EXPLANATORY SUPPLEMENT
Chapront-Touze, M. and Chapront, J. (1983). "The Lunar Ephemeris ELP 2000" Astron.
Astrophys. 124, 50-62.
Chapront-Touze, M. and Chapront, J. (1988). "ELP 2000-85; a Semi-analytical Lunar
Ephemeris Adequate for Historical Times" Astron. Astrophys. 190, 342-352,
Chollet, F. (1984). "Evaluation des corrections de phase dans 1'observation des positions
des planetes" Astron. Astrophys. 139, 215-219.
Clemence, G.M. (1954). "Perturbations of the Five Outer Planets by the Four Inner
Ones" in Astronomical Papers American Ephemeris (U.S. Government Printing Of-
fice, Washington, DC).
Clemence, G.M., Porter, J.G., and Sadler, D.H. (1952). "Aberration in the Lunar Ephem-
eris" Astron. J. 57, 46-47.
Davies, M.E., Abalakin, V.K., Lieske, J.H., Seidelmann, P.K., Sinclair, AT., Sinzi, A.M.,
Smith, B.A., and Tjuflin, Y.S. (1983). "Report of the IAU Working Group on Car-
tographic Coordinates and Rotational Elements of the Planets and Satellites: 1982"
Cel. Mech. 29, 309-321.
Dickey, J.O., Williams, J.G., and Yoder, C.F. (1982). "High-Precision Earth Rotation and
Earth-Moon Dynamics: Lunar Distances and Related Observations" in Proc. IAU
Coll. 63, 0. Calame, Ed. (Reidel, Dordrecht, Holland) p. 209.
Dickey, J.O. and Williams, J.G. (1982). "Geophysical Applications of Lunar Laser Rang-
ing" EOS 63, 301.
Duncombe, R.L. (1969). "Heliocentric Coordinates of Ceres, Pallas Juno, Vesta 1928-
2000" Astronomical Papers American Ephemeris XX part II (U.S. Government Print-
ing Office, Washington, DC),
Eckert, W., Brouwer, D-, and Clemence, G.M. (1951). "Coordinates of the Five Outer
Planets 1953-2060" in Astronomical Papers American Ephemerides XII (U.S. Gov-
ernment Printing Office, Washington, DC).
Eckert, W.J., Walker, M.J., and Eckert, D.D. (1966). "Transformation of the Lunar Co-
ordinates and Orbital Parameters" Astron. J. 71, 314-332.
Eckert, W.J., Van Flandern, T.C., and Wilkins, G.A. (1969). "A Note on the Evaluation
of the Latitude of the Moon" Mon Not. R.A.S. 146, 473-478.
Estabrook, F.B. (1971). "Derivations of Relativistic Lagrangian for n-Body Equations
Containing Relativity Parameters b and c" JPL Internal Communications
Explanatory Supplement to the Astronomical Ephemeris and the American Ephemeris and
Nautical Almanac (H.M. Stationery Office, London), 1961.
Ferrari, A. J., Sinclair, W.S., Sjogren, W.J., Williams, J.G., and Yoder, C.F. (1980). "Geo-
physical Parameters of the Earth-Moon System" J. Geophys. Res. 85, 3939-3951.
Fricke, W. (1971). "A Rediscussion of Newcomb's Determination of Precession" Astron.
Astrophys. 13, 298-308.
Fricke, W. (1982). "Determination of the Equinox and Equator of the FK5" Astron. As-
trophys. 107, L13-L16.
Goldstein, H. (1950). Classical Mechanics (Addison- Wesley, Reading, MA).
Hammond, J.C. and Watts, C.B. (1927). "Results of Observations with the Six-Inch Tran-
sit Circle, 1909-1918" Publ. U.S. Naval Obs. XI.
Hellings, R.W. (1986). "Relativistic Effects in Astronomical Timing Measurements" As-
tron. J. 92, 650-659.
Hughes, J.A. and Scott, D.K. (1982). "Results of Observations Made with the Six-Inch
Transit Circle, 1963-1971" Publ. U.S. Naval Obs. XXIII, part III.
Ianna, P.A. (1974, 1980). Private communication.
5 / ORBITAL EPHEMERIDES OF THE SUN, MOON, AND PLANETS 321
Krogh, p,T. (1972). Lecture Notes in Mathematics, 362 (Springer- Verlag, New York) p.
22.
Laskar, J- (1986). "Secular Terms of Classical Planetary Theories Using the Results of
General Theory" Astron. Astrophys. 157, 59-70.
Lieske, J-H. (1979). "Precession Matrix Based on IAU (1976) System of Astronomical
Constants" Astron. Astrophys. 73, 282-284.
Lieske, J.H., Lederle, T., Fricke, W., and Morando, B. (1977). "Expressions for the Pre-
cession Quantities Based upon the IAU (1976) System of Astronomical Constants"
Astron. Astrophys. 73, 282-284.
Lindegren, L. (1977). "Meridian Observations of Planets with a Photoelectric Multislit
Micrometer" Astron. Astrophys. 57, 55-72.
Lyttleton, R.A., Cain, D.L., and Liu, A.S. (1979). JPL Publication 79-85 (JPL, Pasadena,
CA).
Marsden, B.G., Sekanina, Z., and Yeomans, D.K. (1973). "Comets and Nongravitational
Forces V." Astron. J. 78, 211-115.
Morgan, H.R. (1933). "Results of Observations with the Nine-Inch Transit Circle, 1913-
1926" Publ. U.S. Naval Obs., XIII.
Morgan, H.R. and Scott, F.P. (1948). "Results of Observations made with the Nine-Inch
Transit Circle, 1935-1945" Publ. U.S. Naval Obs., XV, part V.
Mover, T.D. (1971). "Mathematical Formulation of the Double-Precision Orbit Determi-
nation Program (DPODP)" Jet Prop. Lab. Tech. Report #32-1527 (JPL, Pasadena,
CA).
Moyer, T.D. (1981). "Transformation from Proper-Time on Earth to Coordinate Time in
Solar-System Barycenter Spacetime Frame of Reference I, II" Cel. Mech. 23, 33-56,
57-68.
Muhleman, D.O. and Anderson, J.A. (1981). "Solar Wind Electron Densities from Viking
Dual Frequency Radio Measurements" Astrophys. J. 247, 1093-1101.
Mulholland, J.D., Shelus, P.J., and Silverberg, E.C. (1975). "Laser Observations of the
Moon: Normal Points for 1973" Astron. J. 80, 1087-1093.
Murray, C. A. (1989). "The Transformation of Coordinates Between the Systems of B1950.0
and J2000.0, and the Principal Galactic Axes Referred to J2000.0" Astron. Astrophys.
218, 325-329.
Newcomb, S. (1898). "Tables of Motion of the Earth on its Axis and Around the Sun"
in Astronomical Papers American Ephemeris VI, part I (U.S. Government Printing
Office, Washington, DC).
Newcomb, S. (1898). "Tables of the Heliocentric Motion of Mercury" in Astronomical Pa-
pers American Ephemeris VI, part II (U.S. Government Printing Office, Washington,
DC).
Newcomb, S. (1898). "Tables of Heliocentric Motion of Venus" in Astronomical Papers
American Ephemeris VI, part III (U.S. Government Printing Office, Washington,
DC).
Newcomb, S. (1898). "Tables of the Heliocentric Motions of Mars" in Astronomical Papers
American Ephemeris VI, part IV (U.S. Government Printing Office, Washington, DC).
Newhall, XX, Standish, E.M., and Williams, J.G. (1983). "DE102: a Numerically Inte-
grated Ephemeris of the Moon and Planets Spanning Forty-Four Centuries" Astron.
Astrophys. 125, 150-167.
Newhall, XX, Williams, J.G., and Dickey, J.O. (1989). "Earth Rotation from Lunar Laser
Ranging" IAU Symposium 128 (Coolfont, West Virginia).
322
EXPLANATORY SUPPLEMENT
Newhall, XX (1989). "Numerical Representation of Planetary Ephemerides" Cel. Mech
45, 305-310.
Null, G.W. (1976). Private communication.
Null, G.W. (1978). "Programmer Information for the Saturn Satellite Ephemeris 'PARSAT
1108 Software Package" Jet Prop. Lab Interoffice Memo 314.7-252 (using date element
#360) (JPL, Pasadena, CA).
Reasenberg, R.D. and King, R.W. (1979). "The Rotation of Mars" J. Geophys. Res. 84,
no. 811, 6231-6240.
Richter, G.W. and Matzner, R.A. (1983). "Second-Order Contributions to Relativistic
Time Delay in the Parameterized Post-Newtonian Formalism" Phy. Rev D. 28, 3Q07-
3012.
Rivlin, T.J. (1974). The Chebyshev Polynomials (John Wiley & Sons, New York).
Ross, F.E. (1917). "New Elements of M ars" Astronomical Papers American Ephemeris
IX, part II (U.S. Government Printing Office, Washington, DC).
Schmidt, R.E., Corbin, T.E., and Van Flandern, T.C. (1980). "New USNO Zodiacal Star
Catalog" Bull. Amer. Astron. Soc. 12, 740.
Schwan, H. (1977). "Development and Testing of a Method to Derive an Instrumental Sys-
tem of Positions and Proper Motions of Stars" Veroffentlichungen 27, Astronomisches
Rechen-Institut, Heidelberg.
Seidelmann, PK. (1982). "1980 IAU Theory of Nutation: The Final Report of the IAU
Working Group on Nutation" Cel. Mech. 27, 79-106.
Shapiro, I.I. (1964). "Fourth Test of General Relativity" Phys. Rev. Letters 13, 789-791.
Shapiro, I.I., Counselmann, C.C., III, and King, R.W. (1976). "Verification of the Principle
of Equivalence for Massive Bodies" Phys. Rev. Letters 36, 555-558.
Shelus, P.J., Mulholland, J.D., and Silverberg, E.C. (1975). "Laser Observations of the
Moon, Normal Points for 1972" Astron. J. 80, 154-161.
Simon, J.L. and Chapront, J. (1974). "Perturbations du second ordre des planetes Jupiter
et Saturne: Comparaison avec Le Verrier" Astron. Astrophys. 32. 51-64.
Simon, J.L. and Bretagnon, P. (1975). "Perturbations du premier ordre des quatre grosses
planetes: Variations litterales" Astron. Astrophys. 42, 259-263.
Simon, J.L. and Bretagnon, P. (1975). "Resultats des perturbations du premier ordre des
quatre grosses planetes: Variations Litterales" Astron. Astrophys. Suppl 22, 107-160.
Simon, J.L. and Bretagnon. P. (1978). "Perturbations du deuxieme ordre des quatre
grosses planetes: Variations seculaires du demi grand axe" Astron. Astrophys. 69.
369-372.
Simon, J.L. and Bretagnon, P. (1978). "Resultats des perturbations du deuxieme ordre
des quatre grosses planetes" Astron. Astrophys. Suppl. 34, 183-194.
Simon, J.L. and Francou, G. (1982). "Amelioration des theories de Jupiter et Saturne par
analyse harmonique" Astron. Astrophys. 114, 125-130.
Simon, J.L. (1983). "Theorie du mouvement des quatre grosses planetes. Solution TOP82"
Astron. Astrophys. 120, 197-202
Simon, J.L. and Bretagnon, P. (1984). "Theorie du mouvement de Jupiter et Saturne
sur un intervalle de temps de 6000 ans. Solution JASON84" Astron. Astrophys. 138,
169-178.
Standish, E.M. (1973). "The Figure of Mars and its Effect on Radar-ranging" Astron.
Astrophys. 26, 463-466.
Standish, E.M., Keesey, M.S.W., and Newhall, XX (1976). "JPL Development Ephemeris
Number 96" Jet Prop Lab Tech. Report 32-1603 (JPL, Pasadena, CA).
ORBITAL EPHEMERIDES OF THE SUN. MOON. AND PLANETS 323
andish, E.M. (1981). "Two Differing Definitions of the Dynamical Equinox and the
Mean Obliquity" Astron. Astrophys. 101, L17-L18,
andish, E.M. (1982a). "Orientation of the JPL Ephemerides, DE200/LE200, to the
Dynamical Equinox of J2000" Astron. Astrophys. 114, 297-302.
andish, E.M. (1982b). "The JPL Planetary Ephemerides" Cel. Mech. J. 26, 181-186.
Standish, E.M. (1985). "Numerical Planetary and Lunar Ephemerides: Present Status,
Precision and Accuracies" in Relativity in Celestial Mechanics and Astrometry J.
Kovalevsky and V.A. Brumberg, eds. (Reidel, Dordrecht, Holland) pp. 71-83.
Standish, E.M. (1990). "The Observational Basis for JPL's DE200, the Planetary Ephem-
erides of the Astronomical Almanac" Astron. Astrophys. 233, 252-271.
USNO (1984). "The Astronomical Almanac for the Year 1984" (U.S. Government Printing
Office, Washington, DC, H.M. Stationery Office, London).
Watts, OB. and Adams, A.N. (1949). "Results of Observations Made with the Six-Inch
Transit Circle, 1925-1941" Publ. U.S. Naval Obs. XVI, part I.
Watts, C.B., Scott, F.P., and Adams, A.N. (1952). "Results of Observations Made with
the Six-Inch Transit Circle, 1941-1949" Publ. U.S. Naval Obs. XVI, Part III.
Will, CM. (1974). Experimental Gravitation B. Bertotti, ed. (Academic Press, New York).
Williams, J.G., Sinclair, W.S., and Yoder, OF. (1978). "Tidal Acceleration of the Moon"
Geophys. Res. Letters 5, 943-946.
Williams, J.G., Dicke, R.H., Bender, P.L., Alley, CO., Carter W.E., Currie, D.G., Eck-
hardt, D.H., Fluer, J.E., Kaula, W.M., Mulholland, J.D., Plotkin, H.H., Peultney,
S.K., Shelus, P.J., Silverberg, E.C, Sinclair, W.S., Slade, M.A., and Wilkinson, D.T.
(1976). "New Test of the Equivalence Principle from Lunar Laser Ranging" Phys.
Rev. Letters 36, 551-554.
Williams, J.G., Slade, M.A., Eckhardt, D.H., and Kaula, W.M. (1973). "Lunar Physical
Librations and Lunar Laser Ranging" Moon 8, 469-483.
Williams, J.G. (1984). "Determining Asteroid Masses from Perturbations on Mars" Icarus
57, 1-13.
Williams, J.G. and Standish, E.M. (1989). "Dynamical Reference Frames in the Planetary
and Earth-Moon Systems" in Reference Frames J. Kovalevsky, I Mueller, and B.
Kolaczek, eds. (Kluwer, Dordrecht, Holland) 67-90.
Yeomans, D.K. and Chodas, P.W. (1989). "An Asymmetric Outgassing Model for Cometary
Nongravitational Accelerations" Astron. J. 98, 1083-1093.
Yoder, C.F. (1979). "Effects of the Spin-Spin Interaction and the Inelastic Tidal Defor-
mation on the Lunar Physical Librations" in Natural and Artificial Satellite Motion
P. Nacozy and S. Ferrez-Mello, eds. (University of Texas Press, Austin) p. 211.
CHAPTER
Orbital Ephemerides and Rings of Satellites
by James R. Rohde with Andrew Sinclair
6.1 INTRODUCTION
The ephemerides of most satellites in The Astronomical Almanac are calculated
from a set of analytical expressions, termed a "theory," for the motion of the satellite
in an orbit that is basically an ellipse, but subject to various perturbing forces, such
as the oblateness of the planet and the attractions of other satellites and the Sun.
The orbits of the outer satellites of Jupiter are highly perturbed and representations
by theories are not sufficiently accurate. In these cases numerical integrations of the
equations of motion of the satellites are used.
The objective of this chapter is to describe the theory of the motion of each of
the principal satellites in enough detail for search and identification purposes. In
many cases, a description of the complete theory that can be used for the precise
analysis of observations is given, but in cases where the theories contain large
numbers of periodic terms, only an outline of the theory is given, together with
references to where the complete theory can be found.
Data tables for all of the known satellites can be found in Chapter 15. Table
15.9 presents orbital data; Table 15.10 gives physical and photometric data; and
Table 15.11 presents north poles of rotation and prime meridian data.
An alternative to Section F in The Astronomical Almanac, called The Satellite
Almanac, is available on diskette. The Satellite Almanac is a program for MS-
DOS based microcomputers that runs on the IBM PC computers and compatible
machines. The time span of the Satellite Almanac is currently 1990 December 28
to 2000 January 5.
6'11 Orbital Elements
In. this section we use a fairly common notation to define the osculating elements,
formulas, and transformations. The orientation of the orbital plane relative to a
325
326
EXPLANATORY SUPPLEMENT
Z
(Ascending
node of
orbit)
Figure 6.11.1
The orbital elements used to describe the orbital plane relative
to a reference plane and an origin of longitude O.
Figure 6.11.2
An equivalent form of the orbital elements. The end-on view of a great circle on
the celestial sphere is represented as a straight line. Triangles formed by such
lines are spherical triangles, and the formulas of spherical trigonometry apply.
reference plane and origin are described by three elements, which are shown in
Figures 6.11.1 and 6.11.2. These elements are
uj - argument of the pericenter,
f? = longitude of the ascending node,
i = inclination.
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 327
In addition, the size and shape of the orbit are described by the semi-major axis
a and eccentricity e. The position of the satellite along the orbit, measured from
the pericenter, is described by the mean anomaly M or the true anomaly /. The
difference / - M is known as the equation of center. The equation of center can be
approximated to sufficient accuracy for the satellites by
j - M - \ 2e - -e3J s'mM+ -e2sin2Af + — e3sin3M. (6.11-1)
For many satellites the pericenter is ill-defined since e is small, or the node is ill-
defined since /' is small. In the former case it is usual to use the quantity A, rather
than M, and in the latter case the quantity zo, rather than u>, where
A = M + u> + J? (the mean longitude),
S7 = W + j? (the longitude of pericenter) . (6.11-2)
6.12 Secular Perturbations of the Orbit
If the orbit were unperturbed, then all the elements would remain constant except
the mean anomaly M, which would increase at a constant rate n, the mean motion
of the satellite. For a perturbed orbit the same elliptical model can be used, but the
orbital elements, now called osculating elements, are functions of time, typically of
the form of a constant plus periodic terms. In addition, the angular elements can
have a linear (or secular) variation with time. The form of the expressions for the
osculating elements depends on the reference plane chosen. The oblateness of the
planet causes the pole of the orbital plane to precess around the pole of the planet,
and solar perturbations cause it to precess around the pole of the planet's orbit.
Perturbations by other satellites cause precession around the poles of their orbital
planes, but usually this effect is smaller than the other two effects. Because the
satellites' orbital poles are usually close to the pole of the planet, their effect is
primarily an addition to the oblateness effect. The Laplacian plane is the reference
plane about whose axis the satellite's orbit precesses as a result of these two major
effects. The axis of the Laplacian plane is coplanar with and between the polar
and orbital axes of the planet. Hence the node of the orbit regresses around a plane
that lies between the equatorial and orbital planes of the planet, and has a common
line of intersection with the planes. The orbit maintains a constant inclination to
the Laplacian plane (apart from other, usually smaller perturbing effects). The
Laplacian plane is also known as the invariable plane through the planet.
The location of the Laplacian plane is shown in Figure 6.12.1. It divides the
angle between the equatorial and orbital planes into parts / and fj, where
2n2JZfi sin 2(2 " aV2(l - e'2)'112 sin 2/, ,
(6.12-1)
328
EXPLANATORY SUPPLEMENT
Figure 6.12.1
The Laplacian plane
where J2 is the oblateness coefficient of the planet; ro is the equatorial radius of the
planet; n' is the orbital mean motion of the planet; e' is the orbital eccentricity of
the planet; and n is the mean motion of the satellite.
Apart from small short-period perturbations, the motion of the satellite's orbit
relative to the Laplacian plane is such that
7 = constant.
(6.12-2)
e = 0o-K{t- t0),
where 7 is the inclination of the satellite's orbit to the Laplacian plane, and 9 is
the arc reckoned from the intersection of the planet's equatorial and orbital planes
along the Laplacian plane to the node of the satellite's orbit on the Laplacian plane.
However, it is usually more convenient to take a plane other than the Laplacian
plane as the reference plane, as it eases the calculation of perturbations by other
objects, and also eases the transformation of the calculated position to a reference
frame based on the Earth's equator (which is needed for comparison with observa-
tions). If, for example, the orbital plane of the planet in Figure 6.12.1 is taken as
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 329
the reference plane, then the osculating inclination and longitude of the node will
be the angles i and Q. The formulas of spherical trigonometry give
sin / sin(f? - fi\ ) - sin 7 sin 0,
(6.12-3)
sin / co&(Q — J?i) = cos 7 sin i + sin 7 cos i\ cos 9,
where i\, 0\ are the inclination and longitude of the node of the Laplacian plane
relative to the adopted reference plane; and 7 and 9 are defined above. In most
cases 7 is small, and so (SI - S7\) is small as well. Thus the following approximate
formulas are obtained:
i = I, + 7 cos 9,
Q = Qi +7sin0/sini'i. (6.12-^)
Hence the use of a reference plane other than the Laplacian plane introduces a
periodic oscillation into the osculating inclination and longitude of the node. Note
that j'i and J7i are quantities whose values are determined by the physical prop-
erties of the system. The arbitrary constants associated with the inclination and
node are the quantities 7 and the phase of 9. This choice of reference plane also
introduces a periodic variation, equal to the angular difference fi - fi\ -9, into
the longitude of the pericenter and the mean longitude. This periodic variation is
given approximately by
Aw, AA = 7tan ^j'i sinfl. (6.12-5)
For most satellites the oblateness effect on the orbital plane is much larger than
the solar perturbations, and so the Laplacian plane is very close to the equatorial
plane. For a more distant satellite, such as Iapetus, the effects are comparable, so
the Laplacian plane lies about midway between the equatorial and orbital planes.
For the Moon, the solar perturbations dominate, and the Laplacian plane virtually
coincides with the Earth's mean orbital plane — the ecliptic.
The oblateness, solar, and other perturbations also cause secular variations of
the longitude of pericenter (in most cases the pericenter advances around the orbit)
and of the mean longitude of the satellite, so that the mean motion in longitude is
different from (generally greater than) the value derived from Kepler's law for an
unperturbed orbit. It is this overall rate of motion that is quoted as the satellite's
mean motion, and it is thus an observed mean motion. Hence, when using the mean
motion to determine the mass of the planet from Kepler's law, one must allow for
the part of it due to secular perturbations.
330 EXPLANATORY SUPPLEMENT
It is useful to note in this context that the semi-major axis of the orbit is
affected in a similar way. The various perturbing forces cause perturbations of the
eccentricity and the longitude of the pericenter of the form
Ae - A cos M,
(6.12-6)
e Aw - A sin M,
The effect of these terms on the true longitude is zero to first-order. The perturba-
tion of the distance r arising from Ae and Aw is obtained from
r - a(l - eco&M), (6.12-7)
with M = A — w.
Differentiating Equation 6.12-7, we have
Ar = —a cosMAe — ae sinMAzu
(6.12-8)
= -aA.
It is the normal practice to ignore these terms in e and w, and to determine the
value of the semi-major axis from observations that will absorb the constant term.
This must be allowed for when computing the mass of the planet from Kepler's [aw.
6. 1 3 Perturbations due to Cornmensurabilities
There are many instances of pairs of satellites with mean motions, n and n' , having
a relationship of the form
pn = qn', (6.13-1)
where p and q are two small integers. Such a situation is called a commensura-
bility of mean motions or, more simply, an orbital resonance, because it causes
increased mutual perturbations. This resonance effect is caused by a repeated geo-
metric relationship of the positions of the two satellites (their longitudes) relative
to a particular point in their orbits. This point is usually the pericenter of one of
the satellites, but for Mimas- Tethys it is the midpoint of their nodes on the equator
of Saturn. We shall take as an example the case where the commensurability is
2«'-n-* = 0. (6.13-2)
The prime refers to the outer satellite, and w is the secular rate of change of the
longitude of the pericenter of the inner satellite, caused by the oblateness of the
6 / ORBITAL EPHEMER1DES AND RINGS OF SATELLITES 331
planet, other perturbations, and by the resonance itself. As a result the angular
argument
e = 2\'-\-w (6.13-3)
varies very slowly. If 0 either increases or decreases continually, then the motion is
termed circulation, because the position of conjunctions of the satellites circulates
around their orbits, and in this case the mutual perturbations are not particularly
large. In many cases, however, the average rate of change of 8 is zero, and the value
of e oscillates about either 0° or 180°, depending on the type of commensurability.
This type of motion is termed Hbration. In this example, the libration is about 0°,
and we can write 8 as
8 = (A - w) - 2(A - A') = 0° + oscillation. (6.13^4)
At a conjunction of the satellites, A - A' = 0, and so A - w will be close to zero also,
and hence all conjunctions occur near the pericenter of the inner satellite orbit. As
might be expected, this particular type of resonance has a significant perturbing
effect on the longitudes of the two satellites, and on the longitude of pericenter
and eccentricity of the inner satellite. If the libration amplitude is small, then the
libration can be represented adequately by a single periodic term; e.g.,
0 = O°+Bsin(/?f + £). (6.13-5)
If the amplitude is large, e.g., for the Titan-Hyperion and Mimas-Tethys resonances,
then a Fourier series is needed to represent the libration. The amplitude B and
phase e of the libration are the arbitrary constants associated with s and w. The
libration frequency (3 is a function of the masses of the satellites, the closeness of
the commensurability, and the libration amplitude if it is large.
The principal effects of this resonance on the orbit are to cause perturbations
of the longitudes of the satellites with the period of the hbration, and to cause
a forced component of the eccentricity, whose magnitude depends on the mass
of the perturbing satellite and the closeness of the commensurability. This forced
component is usually the largest component of the eccentricity.
The cornmensurabilities that are of most significance in the satellite systems
are listed in Table 6.13.1. In addition many of the features of the rings of Saturn are
associated with cornmensurabilities with satellites. Some of the newly discovered
small satellites of Saturn are in 1:1 cornmensurabilities with the major satellites.
These cornmensurabilities cause the small satellites to follow or precede the major
satellites by 60° in their orbits. The 2:1 cornmensurabilities among the Galilean
satellites Io, Europa, and Ganymede are closely associated with the Laplace com-
mensurability that affects all three satellites, as a result of which the long-term mean
value of (n, - 3«2 + 2n3) is exactly zero, and the angular argument (Aj - 3A2 + 2A3)
332
EXPLANATORY SUPPLEMENT
Table 6.13.1
The Principal Commensurabilities Among Satellites
Argument Satellites Libration Amplitude
2A'
- A-
ZD
Enceladus-Dione
1°5
;
lo-Europa
1°
r^u-i ijpci uaiiyuicuc
3°
2A'
- A -
Enceladus-Dione
C
■
lo-Europa
3°
Euro pa-Ganymede
C
4A'
-2A-
-n-n'
Mimas-Tethys
97°
4A'
-3A -
-zj'
Titan-Hyperion
36°
2A"
-3A'
+ A
lo-Europa-Ganymede
0?07
2A"
-3A'
+ A
Miranda-Ariel-Umbriel
C
Note: Circulation motion is denoted by C. A prime refers to the outer saleflile of a
pair or to the middle satellite of a triplet. Double-primes refer to the outer satellite of
triple commensurabilities.
has a mean value of 180°, and oscillates about this value with amplitude 0?066 and
period 2070 days.
6.14 Long-Period Perturbations by Other Satellites
The pericenters and nodes of satellite orbits move slowly compared with the orbital
speed, and so the relative motion of the pericenters or nodes of a pair of satellites is
slow, and significant perturbations can arise from the arguments m — w' or Q — fl'.
In the planetary system these are called secular perturbations, but they are in fact
of very long period — 25,000 years and longer. In the satellite systems the periods
are much shorter, and the perturbations are usually represented by periodic terms.
The magnitude of this effect depends on the mass of the perturbing satellite and
the size of its eccentricity or inclination. Hence, for example, in the Saturn system
only the effects of Titan are significant. Its effect on the osculating eccentricity e
and longitude of pericenter zu of Rhea are of the form
e cos vj = e0 cos tuD + er cos tzj-t
(6.14-1)
e sin w = ep sin vjp + ef sin wx
where tuT is the longitude of the pericenter of Titan, ef is a forced component
of the eccentricity caused by its resonance with Titan, and ep, zap are the proper
eccentricity and longitude of pericenter of Rhea, so that rop and the phase of zup
are arbitrary constants, and the rate of mp is due principally to the oblateness of
Saturn.
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 333
This effect is particularly noticeable for Rhea, as the forced eccentricity ef is
much larger than the proper eccentricity ep. In this particular case Wj is the proper
longitude of pericenter of Titan, and ef is related to the proper eccentricity and
mass of Titan.
In the satellite systems of Jupiter and Saturn there are several satellites of large
mass compared to that of the other satellites, and so they are liable to cause sizable
forced eccentricities or inclinations on each other's orbits. These forced components
can themselves cause significant forced components on other satellites, and the re-
sulting terms can have frequencies quite different from the natural frequencies of
the system due to the oblateness, etc. To compute these perturbations the eigen-
value theory used for the secular perturbation theory of the planets must be used
(see Brouwer and Clemence, 1961).
6.15 Planetocentric Rectangular Coordinates
The theory of the motion of a satellite gives expressions for the osculating elements
as functions of time. In many cases it is more convenient to represent the position
of a satellite in its orbit in rectangular coordinates rather than osculating elements.
Let the osculating elements of the satellite be a, e, i, A, tu, and Q, Then,
(1) calculate the mean anomaly M from
M= X-m. (6.15-1)
(2) Determine the eccentric anomaly E from Kepler's equation
£-esin£ = M. (6.15-2)
(3) Calculate the coordinates x, y in the orbital plane (where the *-axis is toward
the pericenter) from
x - a(cos E — e),
(6.15-3)
y = a(l -e2)l/2sin£.
Methods for solving Kepler's equation can be found in standard celestial
mechanics texts.
(4) Next calculate the coordinates x\ , y\ , z\ relative to the reference plane and
*-axis direction used for a particular satellite. If, as in Figure 6.11.1, the
orbit is related directly to the reference plane by the orbital elements, then
*i, yi, Z] are given by
= R3(-fi)R,(-0R3(-w)
(6.15-4)
334
EXPLANATORY SUPPLEMENT
Figure 6.15.1
The satellite orbit referred to an intermediate plane, where
P is the pericenter of the satellite orbit and N = nc+AB
where w = w - SI, and R,(a) for i = 1, 2, 3 is a rotation through an
angle a about the current jc-, _v-, or z-axis respectively (see Section 11.41 for
expressions for these rotation matrices). This particular transformation is
given explicitly as
X] = Or cos u - y sin to) cos J? - (jrsinw + ycosw)sini?cos i,
y\ =(xcosw -ysinw)cos J? + (xsinw + y cos w) cos f2 cos (', (6.15-5)
z\ = (x sin u> + y cos w) sin L
In a more complicated case, such as Figure 6.15.1, where the orbit is referred
to an intermediate plane {e.g., the equator of the planet) which is itself referred to
the reference plane and *-axis direction. The coordinates Xt, yi, zi are given by
= R3(-fle)Ri(-ie)R3(-tf+ r?e)R,(-7)R3(-w)
(6.15-6)
where z'e and flt are the inclination and node of the intermediate plane, and 7 and
N are the inclination and node of the orbital plane, as shown in Figure 6.15.1.
g / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 335
Direction of
increasing declination
(to north)
Figure 6.15.2
Coordinates of the satellite relative to the planet: p = position angle, s = angular distance.
£ = Ao cos 8, and r\ - AS . Aa = difference in right ascension (satellite minus planet),
A£ = difference in declination (satellite minus planet), and 6 = declination of planet.
~x2'
"
= Rt(-«)
y\
.22 _
If the coordinates x\, yi, z\ are referred to the ecliptic and equinox of B1950,
then the coordinates Jr2, yi, Zi of the satelhte relative to the center of mass of the
planet, in the frame of the Earth's equator and equinox of B1950, are
(6.15-7)
where e = 23?44578787 is the obliquity of the ecliptic at B1950.
If the coordinates must be referred to the equator and equinox of some epoch
other than B1950 (e.g., of date or of J2000), then precession should be applied to
x2, yi, Z2- The precession matrix from B1950 to J2000 is given in Section 3.591.
We now consider the apparent position of the satellite as seen by an observer
on the Earth. Let £ = Aqcos* and t] = AS be the differential coordinates of the
satellite in right ascension and declination respectively (see Figure 6.15.2). The
differences are in the sense satellite minus center of planet. Then, to first order,
£ = Cv'2 cos a - X2 sin a) I r,
t) - (zi cos 8 — xi sin 6 cos a — y2 sin b sin a) I r,
(6.15-8)
where a and S are the right ascension and declination of the planet; r is the distance
of the planet from the Earth, in the same units as x2l yi, Zi (usually AU); and £ and
V are in radians.
336 EXPLANATORY SUPPLEMENT
The position angle p and angular distance s of the satellite relative to the planet
are given by
tmp = £ / rf:,
9 9 1/9 (6.15-9)
6. 1 6 The Apparent Orbit
The apparent orbit of a satellite on the geocentric celestial sphere is an ellipse that
is the orthogonal projection, in the direction of the line of sight, of the actual orbit
in space. In a circular orbit, the orbital diameter that is perpendicular to the line of
sight is projected into the major axis of the apparent ellipse. At its extremities, the
satellite is at its greatest elongations from the primary. In planning observations
of satellites with short-period orbits, times of greatest elongations are more useful
than differential coordinates.
At times of greatest elongation the satellite and planet center are separated by
an apparent angular distance a I A, where a is the apparent semi-major axis in arc
seconds at a distance of one astronomical unit and A is the geocentric distance of
the primary. The orbital diameter that lies in the plane formed by the line of sight
and the normal to the orbital plane, projects into the minor axis. At the further
extremity of this diameter, the satellite is in superior geocentric conjunction with
the primary, and at the nearer extremity it is at inferior conjunction. The ratio of
the semi-minor axis to the semi-major axis is the absolute value of sinB, where B
is the angle between the line of sight and the plane of the orbit.
On the planetocentric celestial sphere of the primary, the path of the satellite is
a great circle, and the positions of the satellite on the planetocentric sphere at any
time are obtained from the orbital elements. To represent these positions, the same
coordinate systems are adopted as those defined on the geocentric celestial sphere
by the equator and the ecliptic. Because of the mathematically infinite radius of the
celestial sphere, these reference circles are in identically the same positions on the
planetocentric sphere as on the geocentric sphere. The Earth on the planetocentric
sphere is diametrically opposite the geocentric position of the planet, and, therefore,
at right ascension a ± 180° and declination —8, where a and 8 are the geocentric
coordinates of the planet.
Referred to the celestial equator, the position of the great circle that the satellite
describes on the planetocentric sphere is represented by its inclination (J) and the
right ascension (N) of its ascending node, or by the right ascension (N — 90°) and
declination (90° — J) of the pole of the orbit. The major axis of the apparent elliptic
orbit that the satellite describes on the geocentric sphere is parallel to the plane of
this great circle.
In Figure 6.16.1 the following angles are indicated:
N-90
Orbit of
satellite
Pole of
orbit
North pole of
Earth's equator
To Earth
J 80*!
'Equator of Earth
Figure 6.16.1
Planetocentric celestial sphere. The Earth has right ascension, 180° + a, and
declination, -6. Tx = 180" + a and x'x = 1 80° + a - (N - 90°) = a - N - 90°
Figure 6.16.2
Spherical triangle used to compute
U, B, and P
Figure 6.16.3
Spherical triangle used to compute
a and p — P
338
EXPLANATORY SUPPLEMENT
P: The position angle of the minor axis of the apparent orbit, measured from
North. The axis selected is the one pointing toward the pole of the orbit
relative to which the satellite motion is direct.
B: the planetocentric latitude of the Earth relative to the orbital plane of the
satellite; positive toward the pole of the orbit for which motion is direct.
U: the geocentric longitude of the planet measured around the orbit of the satel-
lite, from the ascending node on the Earth's equator. Hence the planetocentric
longitude of the Earth, measured from the same point, is U — 180°.
From spherical triangle QRE shown in Figure 6.16.2 we obtain
cos B sin t/ = + cos / cos <5 sin(o: — AO + sin J sin 6 ,
cos B cos U = + cos 6 cos(q — N),
sin 5 = + sin J cos 8 sin(a — N) — cos J sin 8, (6.16-1)
cos B sin P = - sin / cos(a — AO,
cos B cos P = + sin / sin S s'm(a — N) + cos J cos 6,
The position of the satellite is described by « — the planetocentric longitude of
the satellite measured in the same way as U — and a — the planetocentric angular
displacement of the satellite from the anti-Earth direction; a takes the value 0° to
180°, and has the value 180° when the satellite lies between the planet and the
Earth.
From spherical triangle ETS shown in Figure 6.16.3 we obtain
sin a sin(p — P) — sin(« — [/),
sincrcos(p — P) - sinficos(« - U),
cos a = cosBcos(« — U).
(6.16-2)
Let r be the distance of the satellite from the planet, A be the distance of the planet
from the Earth, As be the distance of the satellite from the Earth, and s be the the
geocentric distance of the satellite from the planet, r, A, and As are measured in
units of length, and s is measured in units of arc.
Then, from the plane triangle formed by the Earth, the planet, and the satellite,
As sins - rsintr,
As cos 5 = r cos (7 + 6.
(6.16-3)
ORBITAL EPHEMERIDES AND RINGS OF SATELLITES
339
North
Figure 6.16.4
Apparent orbit of a satellite of the primary P as projected on the geocentric sphere. This
figure shows the situation when the Earth is north of the orbital plane of the satellite.
With sufficient accuracy we may write:
s- — - sin a = ^- sin a, (6.16-4)
zisinl" A
where a is the apparent radius of the orbit of the satellite at unit distance, ex-
pressed in seconds of arc. Rigorous formulas for p and s in terms of the differential
coordinates of the satellite, and also in terms of the planetocentric coordinates, are
given later.
At greatest elongation, u - tJ = ± 90°, <r = 90°, and S = a I A in position
angle p = P ± 90°. At the extremities of the minor axis of the apparent orbit, the
satellite is at inferior or superior conjunction; a — U is 0° or 180°, a = B, and the
position angle is P or P + 180°. Evidently, when the Earth is north of the orbital
plane, the satellite is at inferior conjunction at the southern extremity of the minor
a^is, and is at superior conjunction at the northern extremity (see Figure 6.16.4);
B is then positive if the motion is direct, negative if retrograde. When the Earth
is south of the orbital plane, superior conjunction is at the southern extremity of
the minor axis, and B is negative in direct motion, positive in retrograde motion.
340 EXPLANATORY SUPPLEMENT
Irrespective, therefore, of whether the motion is direct or retrograde, the position
angle at superior conjunction, when B is positive, is p - P; when B is negative, it is
p = P+ 180°.
The apparent orbit becomes increasingly elliptical as the Earth approaches the
orbital plane, and reduces to a straight line when the Earth is in this plane. As
the Earth passes through the plane and B changes sign, each superior and inferior
conjunction occurs at the opposite extremity of the minor axis from that at which it
previously took place. Similarly, when the Earth is in the plane that is perpendicular
to the orbital plane and contains the celestial pole, the minor axis is exactly in the
north-south direction, and the major axis is exactly east-west. As the Earth passes
through this plane, the extremity of the major axis that formerly was the more
northerly becomes the more southerly.
Only when J is not too greatly different from 0° or 180° is the direction of
the minor axis necessarily nearly enough north and south for the elongations to
be strictly and unambiguously described as eastern and western. When / is in the
neighborhood of 90° , as in the case of the satellites of Uranus and Pluto, the direc-
tion of the minor axis on the celestial sphere ranges from north-south to east-west.
This introduces confusion in the terminology for the elongations, but in general
they are more appropriately regarded as northern and southern than as eastern
and western.
In the ephemerides for finding the apparent distance s and position angle p, the
factor sincr giving the ratio of * to the apparent distance at greatest elongation is
denoted by F, and therefore the apparent distance of the satellite from the primary
is given by
s-Fet/A. (6.16-5)
With Pq denoting an arbitrary fixed integral number of degrees near the value of P
at opposition, the value of p at any time is expressed in the form p\ +p2- Pi is the
sum of the approximate position angle at elongation and the amount of motion in
position angle since elongation. P2, depending on the date, denotes the correction
P — Pq. In calculating F and p\ , the value of the eccentricity of the apparent orbit at
opposition is used; consequently, in the values of s and p that are derived from them,
the effect of the variation of the eccentricity of the apparent orbit is neglected.
6.17 Calculating Tabulated Values
The apparent positions of the satellites are represented by their positions relative
to the primary, expressed either by the apparent angular distance and position
angle, or by the differential spherical (or rectangular) coordinates in right ascension
and declination. The tabular values are corrected for light-time, and are directly
comparable with observations at the tabular times. It is usually sufficiently accurate
to use the light-time from the planet to the Earth. The timescale of the theories
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 341
of the satellites is TDB. The timescale of the ephemerides in The Astronomical
Almanac is UT. The value of TDB — UT adopted for The Astronomical Almanac
from 1990-2000 is approximated by
TDB - UT = 56!059 + 0f2822(y - 1988. 0) + 0!02223(y - 1988. 0)2. (6.17-1)
6.171 Times of Greatest Elongation The method for calculating the inclination,
orbital longitude, and longitude of ascending node with respect to the Earth's
equator, J^ u, and ./V, can vary from satellite to satellite. The descriptions for these
calculations are contained in Sections 6.2 through 6.7. Given that J, u, and N have
been computed for 0h on a given date, and that the right ascension and declination of
the primary, a and 6, are known, U, B, and P are calculated from Equations 6.16-1.
The angle to the nearest elongation is given by
du = 90° — (w — U). (6.171-1)
The time for the satellite to transit this angle is
dt^duln. (6.171-2)
If dt is greater than one day, J, u and N, and then f/, B, and P are redetermined for
0h on the next day. A new dt is then computed. The light-time correction 01? 1 386 1 Zl
is then applied to give the time of elongation. A is the Earth-planet distance.
6.172 Apparent Distance and Position Angle Given J, u, and N on the date
of opposition, P is determined from Equations 6.16-1. P0 is the integral part of
P. Now from /, u, and N at 0h on any date, U, B and P are computed from Equa-
tions 6.16-1. p - P and sine = F are found from Equations 6.16—2. p] is the sum
of p - P and PQ. p2 is the difference P - P0, where P is the value for 0h on the date
in question, not the value at opposition. The determination of a j A is a straight-
forward computation from the semi-major axis, a, and the Earth-planet distance,
A.
6-173 Differential Coordinates To first order, the differential right ascension and
declination in the satellite minus planet are
Aa = s sin p sect (3 + A3),
(6.173-1)
A(3 = j cos/?,
where s is calculated from Equation 6.16-4 and p = p\ +pi-
342
EXPLANATORY SUPPLEMENT
6.18 Notation
Many different notations have been used in the literature for the orbital elements
of the satellites. The notations for the angular elements used in the balance of this
chapter are described below.
For those satellites for which the orbit is referred directly to the ecliptic and
equinox of B1950.0 or J2000.0 the notation is
A = mean longitude,
w = longitude of the pericenter,
Q - longitude of the ascending node,
/ = inclination.
Where an intermediate reference plane is used (usually the equatorial plane of
the planet or the Laplacian plane) the notation used for the angular elements is
L - mean longitude,
P - longitude of the pericenter,
6 - longitude of the ascending node,
7 = inclination.
When the orbital elements are referred to the Earth equator and equinox of
B1950.0 or J2000.0, the notation used for the angular elements is
/ = mean longitude,
7r = longitude of the pericenter,
N = longitude of the ascending node,
J = inclination.
The reason for using the ecliptic as the reference plane for the orbits is for
convenience rather than any good dynamical or observational reason. The planes of
dynamical significance are the equatorial and orbital planes of the planet. The plane
of observational significance is the Earth's equatorial plane. The orbital elements of
the planets are, however, referred to the ecliptic, and so it was probably a convenient
frame in which to calculate solar and planetary perturbations.
6.2 THE SATELLITES OF MARS
Ephemerides of the greatest eastern elongations, and tables for determining the
approximate apparent distance and position angle are given for the satellites of
Mars in The Astronomical Almanac. The ephemerides of the satellites are computed
from elements given by Sinclair (1989).
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES
Phobos
343
a = 6. 26974 x 10-5 au
n = 1128?844556/day
e = 0.0150
7= 1?10
0 = 327?9O-O?43533rf
L = 232?41 + nd + 0?00124y2
/> = 278?96 + 0?43526d
Na = 47?39 - 0°0014y
Ja = 37?27 + 0?0008y
Deimos
wh<?
a = 1.56828 x 10^4au
n = 285? 161 888 /day
e = 0.0004
7 = 1°79
£> = 240?38-0?01801d
L-28?96 + nd-0.27sinA
P= lll?7+0?01798cf
h= 196?55 - Of 01 801
Na =46?37 -0.0014 y
Ja =36?62 + 0?0008y
d = JD- 2441266.5,
These elements are referred to the fixed Laplacian planes of the satellites. The
angular elements are shown in Figure 6.2.1. The orbital elements are defined as
follows; the semi-major axis a; the mean daily motion n; the eccentricity e\ the
inclination 7; the longitude of the ascending node, 9, measured from the ascending
node of the Laplacian plane on the Earth's equator along the Laplacian plane to
the node of the orbit; L, the mean longitude reckoned from the vernal equinox along
the Earth's equator to the ascending node of the Laplacian plane, then along the
Laplacian plane to the ascending node of the orbit, and then along the orbit to
the mean position of the satellite; P, the longitude of the pericenter, measured in
344
EXPLANATORY SUPPLEMENT
Figure 6.2.1
Reference system for Phobos and Deimos. D = position of satellite pericenter, 6 = AB,
L = TA+AB + BC+CD + mean anomaly, P=TA+AB + BC + CD, Na = TE, and
N-TF.
the same mariner as L from the vernal equinox to the pericenter; Na, the longitude
of the ascending node of the Laplacian plane reckoned from the vernal equinox
along the Earth's equator; Ja, the inclination of the Laplacian plane to the Earth's
equator; h, is reckoned along the Laplacian plane from its ascending node on the
orbital plane of Mars to the ascending node of the satellite orbit.
Though the theories for the motion of the satellites contain several periodic
terms, the only periodic perturbation of importance for Earth-based observations
is L = — 0. 27 sin h for Deimos.
J and A/, the inclination and ascending node of the satellite orbit on the Earth's
equator can be found by considering the spherical triangle EFC in Figure 6.2,1. From
spherical trigonometry we have
sin J s'm(N — Na) - sin 7 sin 9,
(6.2-1)
sin J cos(N — Na) = cos 7 sin Ja + sin 7 cos Ja cos 6.
Now find the arc length FC of triangle EFC from
sin J sin FC = sin/, sin 6,
(6.2-2)
sin/ cos FC = cosJa sin 7 + sin/a cos 7 cos 8.
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 345
u follows from
u = FC + L-Na-0 + (f -M), (6.2-3)
where (/—AO is the equation of center that is computed via Equation 6.11-1, where
M = L-P. (6.2-4)
The computation of the tabulated values follows as described in Section 6.17.
6.3 THE SATELLITES OF JUPITER
6.31 The Galilean Satellites
The ephemerides and phenomena of Satellites I-IV given in The Astronomical Al-
manac are based on the theory of Lieske (1978), with constants due to Arlot (1982).
Lieske's theory is a revision of the theory by Sampson (1921). A shortened version
of Lieske's theory is presented below, which will give the positions of the satellites
to about 500km. The constants used below are from Lieske (1987):
107f, = -41279cos(2L, - 2L2);
107v, = -5596 sin(/>3 -P4)- 2198sin(/>, +P2- in, - 2G)
+ 1321 sinOfr) - 1157sin(L, -ILa + Pt)
- 1940sin(L, - 2L2 + P3) - 791 sin(I, - 2L2+P2)
+ 791 sin(L, - 2L2 + P2) + 82363 sin(2L, - 2L2);
107Ci = +7038 sin(ii - 0, + v,) + 1835 sin(L, -<92 + v,);
1076 = -3187008(1^ - Pi) - 1738cos(L2 - Pi)
+ 93748 cosa, -L2);
107v2 = -1158sin(-2/7j+2<?) + 1715sin(-2/7j + 03+# -2G)
- 1846 sin(G) + 2397 sin(/>3 -P4) - 3172 sin(<fr)
- 1993 sin(L2 - Z,3) + 1844sin(L2 - P2)
+ 6394sin(L2 - P3) + 3451 sin(L2 - P4)
+ 4159sin(L, - 2L2 + P4) + 7571 sin(Li -2L2+P3)
- 1491 sin(L, - 2L2+P2) - 185640 sin(Z,[ - L2)
- 803 sin(Z., - Ly) + 915 sin(2Z.j - 2L2);
107C2 = +81575 sin(L2 - 02 + v2) + 4512sin(L2 - 03 + v2)
- 3286sin(L2 - # + v2)\
1076 = -14691 cos(£,3 - Pi) - 1758cos(2L3 - 2L4)
346 EXPLANATORY SUPPLEMENT
+ 6333cos(Z.2 ^£3);
107v3 = ~1488sin(2i7j+2tf) + 411sin(-53 + !?)
+ 346 sin(-04 + <P) - 2338 sin(G)
+ 6558 sin(P3 - Pa) + 523 sin(Pi + P3 - 2/7, - 2G)
+ 314 sin(0,) - 943 sin(L3 - L4)
+ 29387 sin(Z3 - P3) + 15800 sin(L3 - P4)
+ 3218 sin(2L3 - 2L4) + 226 sin(3L3 - 2L4)
- 12038 sin(L2 - L3) - 662 sin(Z4 - 2L2 + P4)
- 1246 sin(L, - 2L2 + P3) + 699 sin(L, - 2L2 + P2)
+ 217sin(Li - 73);
107C3 = -2793 sin(L3 - 92 + y3) + 32387 sin(L3 - 63 + v3)
+ 6871 sin(L3 - 94 + v3) - 16876 sin(L3 - $ + v3);
107£4 = +1656cos(L4 - P3) - 73328 cos(L4 - P4)
+ 182cos(L4 - ilj) - 541 cos(74 + P4 - 2/7, - 2G)
- 269cos(2L4 - 2F4) + 974cos(L3 - L4);
107v4 = -407 sin(-2/,4 + 2«7) + 309 sin(-2/>4 + 84 + &)
- 4840 sin(-2/Jj + 21?) + 2074 sin(-04 +
- 5605 sin(G) - 204 sin(2G)
- 495 sin(5G' - 2G + <f>2) + 234 sin(P4 - /7,)
- 6112 sin(P3 - P4) - 3318 sin(L4 - ^3)
+ 145573 sin(74 - P4) + 178 sin(74 - 77, - G)
- 363 sin(L4 - /Jj) + 1085 sin(74 + P4 - 211 s - 2G)
+ 672 sin(2L4 - 2P4) + 218 sin(2L4 - 2/7, - 2G)
+ 167 sin(2L» - 64 - V) - 142sin(2L4 - 2<P)
+ 148 sin(L3 - 2L4 + P4) - 390 sin(L3 - L4)
- 195 sin(273 - 2L4) + 185 sin(373 - 7L4 + 4^4);
107<4 = +773 sin(L4 - 277, + 9 - 2G + v4)
- 5075 sin(74 - 93 + v4) + 44300 sin(L4 - 94 + v4)
- 76493 sin(L4 + v4); (6.31-1)
where
u =
L2 =
L3 =
106?078590000 + 203?4889553630643 1,
175?733787000+ 101?3747245566245 1,
120?561385500 + 50°31760915340462f,
g / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 347
L4 = 84?455823000 + 2 1?57 1070875 17961 r,
<j>, = 184?415351000 + 0?17356902r,
Pi =82°380231000 + 0° 16102275/,
P2 = 128?960393000 + 0?04645644f,
P-i - 187?550171000 + 0?00712408?,
P4 = 335?30925400O + O°00183939r,
77, = 13?470395000,
01 = 308P365749000 - 0? 13280610?,
02 = 100?438938000 - Of 03261535 1,
63 = 1 18?90892800O - 0?00717678r,
94 = 322?746564000 - 0?00176018r,
& = 316?500101000 - 0?00000248r,
G' = 31?9785280244 + 0?033459733896f,
G = 30?2380210168 + 0?08309256178969453 f,
4>2 =52? 1445966929,
f = JD - 2443000. 5. (6.31-2)
Li, Pj, and 0,- are the mean longitude, proper periapse, and proper node of
satellite i. <pi is the libration phase angle; 77j is the longitude of perihelion of Jupiter;
*P is the longitude of the origin of the coordinates (Jupiter's pole); G' and G are the
mean anomalies of Saturn and Jupiter; 4>2 is the phase angle in solar (A I 7?)3 with
angle 5G' - 2G.
To convert £, v, and £ to rectangular coordinates 1, y, and z in the moving Jovian
equatorial frame, Equations 6.31-3 and 6.31-4 can be used. Satellite subscripts have
been omitted for clarity.
x = a{\ +Ocos(L-i? + v),
y = a{\+Osm(L-$ -v), (6.31-3)
z = a(.
The coefficients a are
ai = 0.00281 9347 au
a2 = 0. 004485872 AU
a3 = 0.0071 55352 AU
a4 = 0.012585436 au.
348 EXPLANATORY SUPPLEMENT
The formula for finding the coordinates in the Earth equatorial frame of 1950.0 is
r = Ri(-e)R3(-J2)R,(-i)R|(-*)R3(-/)f, (6.3M)
where
f = (x,y,z)T
r = (x,y, z)T
e = 23°26,44"84
n = 99? 99754
J = 1?30691
$ = $ -Q
1= 3° 10401.
e is the obliquity of the ecliptic; tt is the longitude of the ascending node of Jupiter;
/ is the inclination of Jupiter's orbit to the ecliptic; / is the inclination of Jupiter's
orbit to the Earth's equator. The rotation matrices R/ are defined in Section 6.15.
The data tabulated in The Astronomical Almanac consist of: the approximate
times of superior geocentric conjunction; the times of the geocentric phenomena;
and the approximate configurations, in graphical form, of the satellites relative to
the disk of Jupiter.
The Universal Time of each superior geocentric conjunction is given for each
satellite to the nearest minute. The phenomena for which times are given are
eclipses, occupations, transits, and shadow transits. The UT of the beginning and
end of each phenomenon (disappearance and reappearance for eclipses and occulta-
tions, ingress and egress for transits and shadow transits) are given to the nearest
minute for all phenomena that are observable. When Jupiter is in opposition, the
shadow may be hidden by the disk and no eclipses can be observed. In general,
eclipses may be observed on the western side of Jupiter before opposition and on
the eastern side after opposition. Before opposition the disappearance only of Satel-
lite I into the shadow may be observed since it is occulted before it emerges from
the shadow; after opposition only the reappearances from the shadow are visible.
The same is true in general of Satellite II, although occasionally both phenomena
can be seen. In the case of Satellites III and IV, both phases of the eclipses are
usually visible except near certain oppositions. Similarly the occultation disappear-
ances and reappearances of a satellite cannot be observed if, at the time concerned,
the satellite is eclipsed. For Satellites I and II there are, in general, cycles of six
phenomena consisting of both phases of both transit and shadow transit, one phase
of the eclipse, and the other phase of the occultation.
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 349
For Satellite IV, none of the phenomena occur when the plane of the orbit of
the satellite — essentially the same as that of Jupiter's equator — is inclined at more
than about 2° to the line from Jupiter to the Earth (for occupations and transits)
or to the Sun (for eclipses and shadow transits).
Owing to the finite disks of the satellites, the phenomena do not take place
instantaneously — the times refer to the center of the disk. In certain favorable
situations of the orbital planes of the satellites relative to the Earth, one satellite
may he eclipsed by the shadow of another, or may occult another. No predictions
of these phenomena are given in The Astronomical Almanac, but predictions are
given in Handbook B.A.A. at the relevant times. A description of the method of
prediction is given by Levin (1934).
The configurations of the four satellites relative to the disk of Jupiter are shown
jn The Astronomical Almanac in graphical form, on the pages facing the tabular
ephemerides of the times of the eclipses and other phemomena.
6.32 The Fifth Satellite, Amalthea
The data given in The Astronomical Almanac consist of the Universal Times of
every 20th eastern elongation. The elongations are computed from the following
circular orbital elements, referred to equator of Jupiter, given by van Woerkom
(1950):
tQ = 1903 September 1.5 UT = JD 2416359.0;
a = 249"55;
7 = 24*. 1;
6> = 82?5-914?62/;
« = 722?63175/day;
Lo = 194?98;
6Lo= -0°113 - 0°0076f + 0?00035f2;
r = (JD-f0)/365.25, (6.32-1)
o is the mean elongation at unit distance; 7, the inclination to the equator of
Jupiter; 8, the longitude of the ascending node; n, the mean motion per solar day;
Lo, the mean longitude at epoch to; and SLo, the correction to mean longitude.
' ~ ?o is reckoned in Julian years. The longitudes are measured in the plane of the
equator of Jupiter, from the ascending node of the mean orbital plane of Jupiter on
the plane of the equator of Jupiter as shown in Figure 6.32.1. From these orbital
elements:
sidereal period 0d49991083 = 1 1^997860,
mean synodic period 0^49996851 = 1 l']999244.
350
EXPLANATORY SUPPLEMENT
Figure 6.32.1
Reference system for Amalthea and Thebe. A = 180° — AB, 9 = 180° + AC,
9 - A = AC, 0 = DC, N = TO. and a0 = TA- 90°
In determining these elements, van Woerkom adopted Souillart's elements of the
equator of Jupiter, but for the calculation of the elongations in The Astronomical
Almanac the right ascension and declination of Jupiter's pole of Davies, et ai (1989)
are used:
a0 = 268?05 - 0?009 T
(6.32-2)
<50 = 64?49 + 0.003r,
where and 6q are referred to the equator and equinox of J2000.0
r*(JD- 2451545. 0) / 36525.
The right ascension, N, of the node of the orbit on the Earth's equator; the
inclination, J, of the orbit to the Earth's equator; and the arc, ip, along the orbit
from the node on the Earth's equator to the node on Jupiter's equator are given by
sin J sin(N — ao) - + cos 8q cos 7 + sin 60 sin 7 cos(0 — A),
sin J cos(A/ - c*o) = - sin 7 sin((? - A),
cos J - + sin 60 cos 7 - cos 60 sin 7 cos(0 - A),
sin J sin rp = + cos6osin(0 — A),
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES
sin J cos ip = + sin S0 sin 7 + cos 60 cos 7 cos(0 -
A),
351
(6.32-3)
in which 6 — A — the arc along Jupiter's equator from its node on the Earth's equator
to the node of the satellite's orbit — is given by
6-A = 219?8 - 2:54057487(/ - f0)
(6.32-4)
= 219°8-2?50405737</,
where d is the number of days from the epoch (JD 2416359.0).
The quantities /V, J, and tp are calculated from the preceding equations at
(V at any date, and are then used to form U (B and P are not required) from
Equations 6.16-1. The orbital longitude, u, measured from the node on the Eart h's
equator, at 0h on that date is derived from
u = Lo + nd — 8 + 4> + SLq,
(6.32-5)
= V> + 112?35 + 2X1003567 d.
Then the times of the elongations on that date are
eastern elongation 0!!033327(C/ + 90° - u) + 01? 138614
western elongation 0')033327( U+ 270° -») + (» 138614
in which A is the geocentric distance of Jupiter. The terms in A are the correc-
tions for light-time; for strict accuracy they should be interpolated to the times of
geometric elongation which the first terms represent.
The calculation of elongation time need be made for only a few of the ephemeris
dates, the others being obtained by means of multiples of the period. For strict
accuracy, the calculation should be repeated with the values of the quantities at
the calculated times instead of at 0"; but for the tabular accuracy of 0?1, this is
unnecessary.
6.33 The Fourteenth Satellite, Thebe
The times of every 10th eastern elongation for Thebe are presented in The As-
tronomical Almanac. The orbital elements are taken from Synnott (1984) and are
referred to Jupiter's equator and autumnal equinox of 1950.0. For the calculation of
the ephemerides, the inclination of the satellite orbit to Jupiter's equator is ignored.
The elements at epoch are
f0 = JD2443937. 817065,
a = 305"95,
352
EXPLANATORY SUPPLEMENT
e = 0.015,
n = 533?7 / day,
7 = Of 8,
Lo = 354?3,
F0 -344?0, (6,33-1 j
where a is the semi-major axis at unit distance; e is the eccentricity; n is the mean
daily motion; 7 is the inclination with respect to Jupiter's equator; £0 is the mean
longitude reckoned from the Jupiter's autumnal equinox; and Pq is the longitude
of perijove reckoned along the Earth's equator to Jupiter's autumnal equinox and
then along Jupiter's equator to the perijove (see Figure 6.32,1).
The times of elongation for Thebe are calculated in the same manner as those
for Amalthea. In this case the orbital longitude («), is derived from
u = Lq + nd - 9 + 4> + if - M), (6.33-2)
where (/ — M) is the equation of center that is computed from Equation 6.11-1,
where
M = Lq + nd - P0. (6.33-3)
6.34 The Sixth through Thirteenth Satellites
Differential right ascensions and declinations of Satellites VI-XIII are presented
in The Astronomical Almanac. The coordinates are referred to the equator and
equinox of J2000.0. They are computed with a variable-order, variable-step-size
numerical integrator developed by Shampine and Gordon (1975). The equations of
motion in x, y and z are:
dh
. 3/2rf
I H
de-
T
fy
2,2
Js
dt1
"i
d2z_
" 3J2rj
1 2fi
JS
dt2 ~
-(3^2- 1)
-(3/i2 - 1)
(5/J2 - 3)
1=1
6
X Xj
r=1
6
yt-y yt_
4
r3-
(6.34-1)
where Mj is the mass of Jupiter; the A/, 's are the masses of the perturbing bodies; rj is
the radius of Jupiter; r-js is Jupiter-satellite distance; Jt>, y,, and Zi are the rectangular
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 353
c0niponents of the satellite-perturbing body distance; is the satellite-perturbing
body distance; rjj is the Jupiter-perturbing body distance. Planetary coordinates
$xe taken from the DE200. The reciprocal masses, in solar mass units, and the Ji
for Jupiter used in integrations are as follows:
Sun 1.0 Mercury 6023600.0
Venus 408523.5 Earth 332946.038
Mars 3098710.0 Jupiter 1047.35
Saturn 3498.0 Uranus 22960.0
Neptune 19314.0 Pluto 130000000.0
J2 of Jupiter 0.01475
The starting planetocentric coordinates and velocities from Rohde (1990) are:
Epoch JD 2447890.5 = 1989 December 30
VI Himalia x
y
z
VII Elara x
y
z
VIII Pasiphae x
y
z
IX Sinope x
y
z
X Lysithea x
y
z
XI Carme x
y
z
XII Ananke x
y
z
XIII Leda x
y
r (au)
-0.0394627736618
-0.0722678670569
-0.0170810682702
-0.0619697457552
-0.0044124460879
+0.0265446659756
-0.0474722144088
-0.1000885430216
-0.0009486933348
-0.0579232252532
+0.1464131750776
+0.0885266909368
+0.0683709404212
-0.0114622412925
-0.0304090473403
-0.1221787634075
-0.1048136497855
-0.0685824702747
+0.1246479748900
-0.0256709692336
-0.0203855454615
+0.0356054434403
-0.0810815597479
+0.0052592795262
v (Au/day)
+0.0013683275790
-0.0002593437765
-0.0010464833518
+0.0004392130964
-0.0020946808809
-0.0003974001515
-0.0010182557362
+0.0011925451174
+0.0009171895371
+0.0011014867506
+0.0002877970106
-0.0003615702389
+0.0006111658968
+0.0012863404782
+0.001 3466750538
-0,0007075200606
+0.0009285971336
+0.0001468741809
-0.0006562509816
-0.0013514562258
+0.0002939680931
+0.0013645601641
+0.0006286869081
+0.0005799993600
The differential coordinates are computed from Equations 6.15-8.
354 EXPLANATORY SUPPLEMENT
6.4 THE RINGS AND SATELLITES OF SATURN
6.41 The Rings of Saturn
The rings of Saturn lie in its equatorial plane. The inclination and ascending node
of this plane on the Earth's equator are determined from the north pole of Saturn
as denned by Davies et al. (1989). The dimensions of the outer ring and the factors
for calculating the relative dimensions of the rings are from Esposito et al. (1984).
The outer, inner, and dusky rings correspond to the A, B, and C rings, respectively.
The Cassini division lies between the inner and outer rings.
The ephemeris of the rings contains the following quantities that determine the
Saturnicentric positions of the Earth and Sun referred to the plane of the rings,
upon which the appearance of the rings depends:
U = the geocentric longitude of Saturn, measured in the plane of the rings
eastward from its ascending node on the mean equator of the Earth;
the Saturnicentric longitude of the Earth, measured in the same way, is
U+ 180°.
B - the Saturnicentric latitude of the Earth referred to the plane of the rings,
positive toward the north; when B is positive, the visible surface of the
rings is the northern surface.
P = the geocentric position angle of the northern semi-minor axis of the
apparent ellipse of the rings, measured from the north toward the east.
U' = the heliocentric longitude of Saturn, measured in the plane of the rings
eastward from its ascending node on the ecliptic; the Saturnicentric lon-
gitude of the Sun, measured in the same way, is U' + 180°.
B' = the Saturnicentric latitude of the Sun referred to the plane of the rings,
positive toward the north; when B' is positive, the northern surface of
the rings is the illuminated surface.
P' = the heliocentric position angle of the northern semi-minor axis of the
rings on the heliocentric celestial sphere, measured eastward from the
circle of latitude through Saturn.
The right ascension and declination of the north pole of Saturn are according
to Davies et al. (1989),
a, = 40?58 - Of 036 T,
(6.41-D
<5j = 83?54-0?OO4 7\
where a} and 6j are referred to the equator and equinox of J2000.0 (T = (JD -
245 1545.0)/ 36525).
Adding the variations due to precession, calculated by the formulas in Sec-
tion 3.21, gives the elements referred to the ecliptic and mean equinox of date; from
g / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 355
them, the inclination, J, to the mean equator of date and the right ascension, A/, of
the ascending node measured from the mean equinox of date are obtained from
J = 90° - S0
(6.41-2)
N = a0 + 90?
The arc <j> from the ascending node of the ring plane on the Earth's equator along
the ring plane to its ascending node on the ecliptic is obtained from
sin N cot <j> - cos N cose - sin scot/, (6.41-3)
where e is the mean obliquity of date. From the elements referred to the equator of
the Earth, and from the geocentric equatorial coordinates of Saturn, the ephemeris
of V, B, and P is calculated by Equations 6.16-1. These quantities are defined with
reference to the ring plane in the same way as with reference to the orbital plane
of a satellite; but the effect of nutation must first be removed from the apparent
right ascension and declination of Saturn, to refer the position of the planet to the
same equinox as the elements of the reference plane.
U' , B', and P' may be obtained by formulas exactly analogous to Equations 6.16-1
from the ecliptic elements of the plane of the rings and the heliocentric longitude,
/, and latitude, b, of Saturn, referred to the mean equinox of date:
cos B' sin P'
- - sin i cos(/ - Q),
cos B' cos P'
- + cos / cos b + sin /' sin b sin(/ —
Q%
sinZJ'
= — cos i sin b + sin /' cos b sin(/ -
f2),
cos B' sin U'
= + sin i sin b + cos /' cos b sin(/ -
m,
cosB' cos U'
= + cos b cos(/ - Q).
(6.41-*)
The rings become invisible when
(a) The ring-plane passes through the Sun, since neither side of the rings is
then illuminated.
(b) The ring-plane passes between the Sun and the Earth, since the unillumi-
nated side of the rings is facing the Earth.
(c) The ring-plane passes through the Earth, since the rings are too thin to be
visible edge on.
Twice during each revolution of Saturn around the Sun — near the times when
the ring-plane passes through the Sun — the Earth usually crosses the ring-plane
three times, though occasionally it crosses only once. On 1995 May 22, the Earth
356 EXPLANATORY SUPPLEMENT
again crosses the ring-plane. The rings will remain continuously invisible until 1995
August 11 while the Earth is on the unilluminated side. The Earth crosses the
ring-plane for a third time on 1996 February 11. The last occasion on which only
one passage of the Earth through the ring-plane occurred was in 1950.
6.42 The Satellites of Saturn
The Astronomical Almanac presents times of greatest eastern elongation for Satel-
lites I-V, times of elongations and conjunctions for Satellites VI-VIH, apparent
distances and position angles for Satellites I- VIII, orbital elements referred to Sat-
urn's equator for Satellites I- VIII, and differential right ascensions and declinations
for Satellites VII-IX. The orbital elements published are as follows: L, the mean
longitude reckoned from the ascending node of Saturn's equator on the Earth's
equator along Saturn's equator to the ascending node of the satellite's orbit then
along the orbit to the satellite; M, the mean anomaly; $, the longitude of the as-
cending node of the orbit on Saturn's equator measured from the ascending node
of Saturn's equator on the Earth's equator; 7, the inclination of the orbital plane
to Saturn's equator; e, the eccentricity; and a, the semi-major axis.
Eclipses, occultations, transits, and shadow-transits of the satellites occur dur-
ing a limited period each time the Earth passes through the plane of the rings.
Methods for calculating these phenomena are described by Comrie (1934). See also
Taylor (1951).
Ephemerides are not provided for Satellites X-XVII though orbital, physical,
and photometric data are listed on pages F2 and F3 of The Astronomical Almanac.
Janus was discovered in 1967. Epimetheus, Helene, Telesto, and Calypso satellites
were discovered by ground-based observations during 1980. Atlas, Prometheus, and
Pandora were discovered by Voyager. Many of the satellites are coorbital. Janus
and Epimetheus follow horseshoe paths in a reference frame rotating at their mean
orbital speed. Telesto and Calypso orbit at the L4 and L5 Lagrangian points of
Tethys. Helene is near the L4 Lagrangian point of Dione. Pandora and Prometheus
are the shepherd satellites of the F-ring.
6.421 Mimas, Enceladus, Tethys, and Dione The orbital theories for Mimas, Ence-
ladus, Tethys, and Dione are based on Kozai (1957). The values of the elements for
Mimas, Enceladus, Tethys, and Dione are from Taylor and Shen (1988), with mean
motion and secular rates for Mimas and Enceladus from Kozai and corresponding
quantities for Tethys and Dione from Garcia (1972). The elements of Mimas, Ence-
ladus, Tethys, and Dione are referred to equator of Saturn (see Figure 6.421.1). The
only significant periodic perturbations on these satellites are the mutual perturba-
tions of Mimas and Tethys, and those of Enceladus and Dione. In each case, they
are caused by a close 2:1 commensurability of mean motions. These perturbations
affect the longitude and are denoted by AL.
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES
357
Figure 6.421.1
Reference plane for Mimas, Enceladus, Tethys and Dione. 9\ = Tfi + BC,
P = TB + BC + CD,N= and L\ = Tfl + BC + CD + mean anomaly.
The orbital elements of Mimas are as follows:
d= JD - 2411093.0,
t = dj 365. 25,
T= 5?0616[((JD -2433282.423)/ 365.25)+ 1950.0- 1866.06],
a = 0.00124171 AU,
n = 381?994516/day,
e = 0.01986,
7 = 1?570,
61, = 49?4- 365?025/,
h = 128?839 + nd - 43°415 sin T - 0?714 sin 3 T
- 0?020 sin 5 T,
P= 107?0 + 365?560f, (6.421-1)
where a is the semi-major axis; n the daily mean motion; e the eccentricity; 7 the
inclination of the orbital plane to Saturn's equator; 9{ the longitude of the ascending
n°de; ti the mean longitude; and P the longitude of the pericenter.
358 EXPLANATORY SUPPLEMENT
The orbital elements of Enceladus are given by:
d = JD - 2411093.0,
r = cf/365. 25,
a = 0.00 15 8935 au,
« = 262?73 19052 /day,
e = 0.00532,
7 = 0?036,
B\ = 145° - 152?7f,
hi = 200? 155 + nd + 1 5'.38 sin(59?4 + 32°73 /)
+ 13'.04sin(119?2 + 93?18r),
P = 312?7 + 123?42^, (6.421-2)
where the variables are defined in the same manner as for Mimas.
The orbital elements of Tefhys are as follows:
<f = JD- 241 1093.0,
f = <*/365.25,
T = 5?0616[((JD - 2433282.423) / 365. 25) + 1950. 0 - 1866. 06],
a = 0.00197069 AU,
n = 190?697920278 / day,
e = 0. 000212,
7 = 1? 1 121,
0, = 111?41 -72?24754f,
L\ = 284?9982 + nd + 2?0751 sin T + 0?0341 sin 3 T
+ 0?0010sin5 7',
P = 97°+72°29/, (6.421-3)
where the variables are defined in the same manner as for Mimas.
The orbital elements of Dione are given by:
d = JD -2411093.0,
t = dl 365.25,
<j = 0. 00252413 au,
n= 131?534920026/day,
e = 0. 001715,
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 359
7 = 0.° 0289,
<?i = 228° -30?6197r,
U = 255f 1 183 + nd - 0'.88 sin(59?4 + 32°73 f)
-0'.75 sin(119°2 + 93?18r),
P = 173?6 + 30?8381 1, (6.421-4)
where the variables are defined in the same manner as for Mimas.
To determine the times of eastern elongation we must first find the orbital
longitude, u\ the longitude of the ascending node N of the orbit on the earth's
equator; and the inclination of the orbit to the Earth's equator, J. From spherical
trigonometry we have from triangle BFC in Figure 6.421.1,
sin i sin BF = sin 7 sin(#j — Qe),
(6.421-5)
sin i cos BF - cos 7 sin ie + sin 7 cos /„ cos(#i - Qs),
which can be solved for BF and i. The arc length FC can be found from the equation
sin i sin FC = sin(0] - J7e) sin ie. (6.421-6)
Similarly, from the triangle TEF we have
sin J sin N = sin i sin( J?e + BF),
sin/cos/V = cosjsine + sini'cos£COs(/2e + BF),
(6.421-7)
which can be solved for J and ./V. Then the arc EF can be found from
sin J sin EF = sin(fie + BF) sin g (6.42 1-8)
u then follows from
u = P-9l+EF + FC+ true anomaly. (6.42 1-9)
The tabulated orbital elements 7, a, and e are listed above. The remaining elements
must be computed. From the spherical triangle TAB in Figure 6.421,1 we have
sin./e sin AB = sine sin Qt, , a„,
(6.421-10)
sin Je cos - cos e sin ie + sin g cos ie cos i?e,
which can be solved for AB.
360
Then
e = ex- J7e+AB,
L = Li — Q + 6,
M=Li - P.
EXPLANATORY SUPPLEMENT
(6.421-H)
6.422 Rhea, Titan, Hyperion, and lapetus The orbital theories for Rhea and Ti-
tan are from Sinclair (1977). The orbital elements are from Taylor and Shen (1988)
with the mean motions and secular rates used from Garcia (1972), The theory and
orbital elements for Hyperion are from Taylor (1984). The theory for the motion
of lapetus is taken from Sinclair (1974) with some additional periodic terms from
Harper et ai (1988). The orbital elements for lapetus are from Taylor and Shen
(1988).
Because of the oblateness of Saturn and solar perturbations, the orbit plane of
Rhea precesses at constant inclination around the Laplacian plane of Rhea. Because
we are using the ecliptic as the reference plane, this motion around the Laplacian
plane appears as periodic variations in A, vj, J?, and i . The osculating elements are
given by:
a = 0. 00352400au,
n = 79?69004007007 day,
7o = 0?3305,
e sin tu = 0. 000210 sin tt + 0. 00100 sin wT,
ecosro = 0.000210cos7T + 0.00100 cos tuT,
A = 359f4727 + «^
+ k sin 70 tan sin(356?87 - 10?2077 r),
i = ie - 0?0455
+ «cos(356f 87 - 10°2077 1) + 0?0201 cosJVT,
n = J2e - 0?0078 + [Ksin70sin(356?87 - 10°2077/)
+ 0?0201 sin Nj] I sin ie, (6.422-1)
where
7r = 305° + 10?2077/,
wT - 276?49 + 0?5219(JD - 241 1368. 0) / 365. 25,
iVT = 44?5 - 0?5219(JD - 241 1368. 0) / 365. 25,
K = 57P29578,
d=JD- 241 1093.0,
ORBITAL EPHEMERIDES AND RINGS OF SATELLITES
361
Figure 6.422.1
Reference system for Rhea, Titan, Hyperion, lapetus, and Phoebe,
to = TF + FD and A = TF + FD + mean anomaly.
t = di 365.25. (6.422-2)
Here a is the semi-major axis; n is the daily mean motion; e is the eccentricity; w
and A are the longitude of pericenter and the mean longitude, respectively, reckoned
from the vernal equinox along the ecliptic to the ascending node and thence along
the orbit; i and D, are the inclination and ascending node of the orbit with respect
to the ecliptic of 1950.0. The angular elements are shown in Figure 6,422.1.
The coefficient 0.00100 in the expressions for esinnj and ecosn? is a forced
eccentricity due to Titan. The algebraic expression for this is given in Sinclair
(1977). The inclination and node of Rhea's Laplacian plane relative to the ecliptic
(1950.0) are i = ie - 0?0455 and J2 = J?e — 0°0078.
There are a number of significant solar perturbations on Titan's orbit and, as
for Rhea, A, J7, and i are affected by periodic variations due to the reference plane
nc>t being the Laplacian plane. The solar perturbations are functions of the mean
dements of the apparent orbit of the Sun about Saturn. The solar elements needed
are As, J?s, and js (which are defined in Figure 6.422.2), and /s, the mean anomaly.
362
EXPLANATORY SUPPLEMENT
Figure 6.422.2
Angles needed to evaluate solar perturbations on Titan.
na = TD, 0 = TA+AB, w, = TD + DC, and Xs = TA+AE.
Approximate values of i, f2, and W are also needed, and these are denoted by ia,
J?a, and C7a. These quantities are given by the following expressions:
T = (JD - 2415020. 0) / 36525
ls = 175?4762+ 1221 ?55 15 7
i, = 2?4891 +0.° 002435 T
J7S = 113?350-0?2597 7"
As = 267?2635+ 1222? 1136 T-
/ = (JD- 241 1368.0)/ 365. 25
k m 57?29578
t, = ie - Of 6204+ Ksin0f2990cos(41°28 - 0?5219/)
na = Qe - 0? 1418 + Ksin 0?2990sin(41?28 - 0f5219 f) / sin/e
CTa = 275?837 + 0?5219r (6.422-3)
6 / ORBITAL EPHEMER1DES AND RINGS OF SATELLITES 363
The quantities (ie - 0?6204) and (l?e - Of 1418) define the inclination and node of
the Laplacian plane of Titan. The auxiliary angles — 0, T, g, and Ls — are given
by:
cos r = cos t's cos j'a + sin t's sin ia cos(J?a - J?s),
sin F sin <P = sin t's sin(i?a — J2S),
sin T cos = cos is sin ia — sin is cos ia cos(J2a - J?s),
sin T sin(# — = sin z'a s'm(Q-d - fis),
sin T cos(9 — J?s) = — sin is cos i"a
+ cos is sin t'a cos(J?a - f2s),
Ls = As - (0 - f2s) - f2s,
g = ma- a^-V. (6.422^)
The osculating elements of Titan are given by:
a = 0. 00816765 au,
n = 22?57697385/day,
e = 0.02881 5 -0.000184 cos 2g + 0.000073 cos 2(L* - g),
w = w& + k[0. 00630 sin 2g + 0. 00250 sin 2(L5 - g)],
A = 261?3121 +«rf+ «[sin70tan ^/esin(41?28
- 0?52 19 f) - 0. 0001 76 sin J, - 0. 000215 sin 2L,
+ 0. 000057 sin(2Ls + !?)],
i = ia + 0. 000232k cos(2Ls + 9),
Q = Qa + 0. 000503k sin(2Ls + P\ (6.422-5)
where d= JD - 2411368.0; f = 4/365.25; and k = 57? 29578.
The algebraic expressions for the coefficients of these periodic terms are given
in Sinclair (1977).
The osculating elements of Hyperion's orbit are given by:
a - 0. 0099040 - 0. 00003422 cos t au,
«= 16?9199514/day,
e = 0. 10441 - 0. 00401 cos r + 0. 00009 cos(C - r)
+ 0. 02321 cos C - 0. 00009 cos(C + r) - 0. 001 10 cos 2C
+ 0. 00013 cos(3 1 ?9 + 61 ?7524 7"),
i = ie - 0?747 + 0?6200cos(105?31 - 2?392 70
364 EXPLANATORY SUPPLEMENT
+ 0°315cos(38?73 - Of 5353 r)
-0?018cos(i3o+24°44r),
n = Qe + ((-0P061 + 0?6200sin(105°31 - 2?392D
+ 0°315 sin(38°73 - 0°5353 71
- 0?018 sin(13° + 24°44 7)) I sin(ie - 0. 747),
A= 176°7481 + wj + 0°1507sin(105?31 -2?3927)
+ 9?089 sin r + 0?007 sin 2r - 0°014 sin 3t
+ 0? 192 sin(C - t) - 0?091 sin C + Of 21 1 sin(C + r)
- 0?013 sin(176° + 12°22 T) + 0?017 sin(8° + 24?44 T),
tu = 69?993 - 18?6702r + 0?1507sin(105?31 -2°392T)
-0?47sinr- 13?36 sin C + 2° 16sin2C
+ 0?07sin(31?9 + 61o7524D, (6.422-6)
where
d = 3D- 2415020. 0,
71 - ( JD - 2433282. 42345905) / 365. 2422 + 50. 0,
t = 93?13 + 0°562039rf,
C= 148?72- 19° 184 T.
Here again the elements are reckoned in the same way as for Rhea.
The orbital plane of Iapetus is inclined at about 8° to its Laplacian plane,
which is itself inclined at about 15° to the equatorial plane of Saturn, and 12° to
the orbital plane of Saturn. As a result, the motion of the orbital plane of Iapetus
is quite complicated, and is described by cubic polynomials in time for ; and Q-
The osculating elements are computed as follows:
/= JD - 2409786.0,
tc = tl 36525,
T = (JD - 241 5020. 0) / 36525.
The elements computed from constant and secular terms are:
a = 0. 02380984 au,
n = 4f53795711 /day,
e = 0. 0288 1 84 + 0. 000575 tc,
i = 18?45959-0°9555/c - 0? 0720 % +0.0054 f3,
6 j ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 365
A = 76° 19854 +
ro = 352?905 + ll°65fc,
n - 143? 1294 - 3f 797 tc + 0? 1 1 6 t2c + Of 008 i* . (6.422-7)
The auxiliary angles needed to compute periodic terms are calculated as follows:
# = 4°367-0?195 7\
0 = 146?819- 3°918 7/,
As = 267?263 + 1222. 114 T,
H7s = 91°796 + 0f562r,
AT = 261°319 + 22?576974(JD - 2411368.0),
H7T = 277° 102 + 0°001389(JD - 241 1368. 0),
<£ = 60°470+l°521 T,
§ = 205?055 - 2?091 T. (6.422-8)
From these we compute the following:
/ = A - m,
lT = AT — wj,
g = zu — n — 6,
gj — Wi —
g] = w - n - (6.422-9)
Some of the expressions for the perturbations involve e and i. For these the following
expressions are used.
e = 0.0288184 + 0?000575?c, ]m
(0.4Z2-1U)
i = 18?45959 - 0?9555 te - 0?0720f? + 0?0054fc3.
The following perturbations are to be added to the elements
Aa = 10"5a[7. 87cos(2/ + 2g - % - 2gs) + 98. 79cos(/ + gi - h - gt)]
Ae = 10_5[- 140. 97 cos{gi - gT) + 37. 33 cos(2/s + 2gs - 2g)
+ 1 1 . 80 cos(/ + 2g - 21s -2gs) + 24. 08 cos I + 28. 49 cos(2/ + gi - h - gt)
+ 61.90cos(/T + gT -gi) + 0.496cos(3/s + 2gs - 2g)]
366 EXPLANATORY SUPPLEMENT
A/ = Of 04204 cos(2/s + 2gs + 0) + Of 00235 cos(/ + g\ + lT + gl + <f>)
+ Of 00360 cos(/ + g] - h - gj + (p) + Of 0005 cos(4/s + 2gs)
+ 0f0058cos(3/s + 2gs) - 0f0024cos(/s + 2gs)
AA = -Of 04299 sin(/ + g] - lT - gT) - Of 00789 sin(2/ + 2g - 2/5 - 2gs)
- 0f06312 simJi - 0f00295 sin2/s - 0°02231 sin(2/s + 2gs)
+ Of 00650 sin(2/s + 2gs + 0)
Atu = [Of 08077 sin(gi - gT) + 0f02139 sin(2/s + 2gs - 2g)
- 0°00676 sin(/ + 2/g - 2/s - 2gs) + OfO 1380 sin/
+ 0?01632sin(2/ + gi - /T -gT) + Of 03547 sin(/T + gj - gj)
+ Of 00028 sin(3/s + 2gs - 2g)] I e
AH - [0?04204 sin(2/s + 2g% + $) + 0?00235 sin(/ + gi + lT + gT + <j>)
+ 0?00358 sin(/ + g\ - h - 8t + 4>) - Of 0006 sin 2/s
+ 0?0003 sin(4/s + 2gs) + Of 0028 sin(3/s + 2gs)
- 0?0012 sin(/s + 2gs) - Of 0142 sin /J / sin i, (6.422^1 1)
To determine J, u, and N, we can use the spherical triangle TEF in Fig-
ure 6.422.1 to obtain
sin / sin N = sin i sin Q,
(6.422-12)
sin J cos N = cos sin e + sin i cos e cos Q,
which can be solved for N and J. The arc <j> can be found from the equation
sin J sin <j> = sin Q sine. (6.422-13)
u then follows from
u-w — Q + 4> + true anomaly (6.422-14)
The tabulated orbital elements a and e are shown previously. The remaining
elements are computed as follows. To calculate 6 we first find the arc AC in Fig-
ure 6.421.1. From spherical trigonometry we have for the triangle AEC:
sm(N -Ne)cot6 = cos(N-Ne)cosJe-smJecotJ, (6.422-15)
which can be solved for 6. M is given by
M = X-Q.
(6.422-16)
6 / ORB|TAL EPHEMERIDES AND RINGS OF SATELLITES
^ be determined by the formula
367
siii7sin0 = sin7sin(/V - Ne). (6.422-17)
I js found by first finding the arc length FC from
sin FC sin 7 * sin( Q - Qt) sin ( . (6.422-1 8)
L is then given by
L=X- n - FC + 6. (6.422-19)
The differential coordinates of Hyperion and Iapetus are computed as follows:
the rectangular coordinates x and y of the satellite in the orbital plane are found
from Equations 6.15 3. These coordinates axe then transformed to the Earth equa-
tor and equinox by Equations 6.15-4 and 6.15-5. The differential coordinates £ and
77 are then computed through Equations 6.15-8.
6.423 Phoebe The orbital elements used in The Astronomical Almanac are those
determined by Zadunaisky (1954). The ephemerides consist of differential right
ascensions and declinations given at 2-day intervals. The elements are referred to
the ecliptic and equinox of 1950.0. The values of the elements are as follows:
a = 0. 0865752 AU,
A = 277f 872 - Of 6541068 1,
e-0. 16326,
m = 280f 165 -Of 19586 T,
i= 173f949 -0f020F,
fi = 245?998 -Of41353 7\ (6.423-1)
where
/ = JD- 2433282. 5,
r=f/365.25.
« is the semi-major axis; A, the mean longitude measured from the equinox along
the ecliptic to the ascending node then along the orbit; e, the eccentricity; w, the
iongitude of the pericenter reckoned from the equinox along the ecliptic to the
ascending node then along the orbit to the pericenter; /, the inclination of the orbit
to the ecliptic; and Q, the longitude of the ascending node of the orbit on the
ecliptic reckoned from the equinox.
368 EXPLANATORY SUPPLEMENT
To compute the differential right ascension and declination, the x and y coor-
dinates in the orbital plane are found using Equation 6.15-3, where
M=X-w. (6.423-2)
The coordinates are transformed to the Earth equator and equinox using Equa-
tions 6.15-4 and 6.15-5. Finally, £ and r\ are computed from Equation 6.15-8.
6.5 THE RINGS AND SATELLITES OF URANUS
Orbital elements are given for the rings of Uranus as determined by Elliot et al.
(1981). Ephemerides are given for the elongations and for the apparent distance
and position angle, of Satellites I through V; Ariel, Umbriel, Titania, Oberon, and
Miranda. The orbital elements of these satellites are from Laskar and Jacobson
(1987). The theory is developed in elliptical elements. Secular terms and periodic
terms with amplitudes greater than 0?05 at opposition are quoted below,
Miranda
mIMxi =0.075 x 10-5,
n = 4. 44352267 - 136. 65 cos(0. 621 T+5. 89) radians/day,
L = -0. 23805158 + 4. 44519055 T
+ 0. 02547217 sin(-2. 18167 x 10_4T + 1.32)
- 0. 0030883 1 sin(-4. 36336 x 10~4T + 2. 64)
-3.181 x 10_4sin(-6. 54502 x 10^4T + 3. 97) radians,
Z= 1.31238 x 10"3 exp 5273 x KT4T + 0.61)
- 1.2331 x 10~4 expi(0. 08606 T + 0, 15)
- 1.9410X 10^4exp/(0.709T+6.04),
C = 0. 03787171 expi(-l. 54449 X 10~4T+5.70). (6.5-1)
Ariel
m/Mv = 1.49 x 10-5,
n = 2. 49254257 radians/day,
1 = 3. 09804641 + 2. 49295252 T
- 1. 86050 x 10-3sin(-2. 18167 x \0~AT + 1.32)
+ 2. 1999 x 10"4 sin(-4. 36336 x 10 4r+ 2. 64) radians,
i= 1.18763 x 10^3exp/(4. 727824 x 10_5r+2.41)
/ ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 369
+ 8.6159x 10_4expi(2. 179316x 10-5T + 2.07),
C - 3. 5825 x 10 4 exp i(-4. 782474 x 10_5T + 0. 40)
+ 2. 9008 x 10~4 exp t(-2. 156628 x 10"5T + 0. 59). (6.5-2)
Umbriel
m/Mv = 1.45 x 10~5,
n - 1. 51595490 radians/day,
L = 2. 28540169 + 1.51614811 T
+ 6.6057 X 10_4sin(-2. 18167 x 10_4T+ 1.32) radians,
Z = -2.2795 X 10 4 exp i(4. 727824 x 10^7+2. 41)
+ 3. 90469 x 10-3 exp i(2. 179132 x 10-57 + 2,07)
+ 3.0917 x 10"4 exp t'(l. 580524 x 10_5T+0.74)
+ 2.2192 x 10-4 exp i(2. 9363068 x 10-6T + 0.43)
+ 5. 4923 x 10~4 exp i'(-0. 01 157 T + 5. 7 1),
C= 1.11336 x 10~3 exp f(-2. 156628 T + 0. 59)
3.5014 x 10-4exp/(-l. 401373 xl0-5T+ 1.75). (6.5-3)
Titania
m/Mu = 3.97x 10"5,
n = 0. 72166316 radians/day,
L = 0. 85635879 + 0. 72171851 Tradians,
2 = 9.3281 x 10~4 exp 580524 x 10_5r+0.74)
+ 1. 12089 x 10~3 exp i(2. 9363068 X 10_6T + 0.43)
+ 7. 9343 x 10~4 exp i(-6. 9008 x 10_3T+ 1. 82),
C = 6. 8572 x 10-4 exp /(- 1 . 401373 x 10_5T + 1 . 75)
+ 3. 7832 x 10-4 exp 9713918 x 10-6T + 4.21). (6.5—4)
Oberon
m!Mv = 3.45 x 10-5,
n = 0.46658054 radians/day,
L = -0.91559180 + 0. 466692 1 2 T radians,
370 EXPLANATORY SUPPLEMENT
z= -7.5868 x 1 (T4expi'{l. 580524 x Kr57" + 0.74)
+ 1. 39734 x 10-3 exp f(2. 9363068 x I0_67; + 0.43)
- 9.8726 x 10-4 exp i(-6. 9008 x 10_3r+ 1. 82),
C = -5.9633 x 10~4 exp /(-l. 401373 x IQ~5T+ 1.75)
+ 4.5169 x 10_4expi(-l. 9713918 x 10"67" + 4.21), (6.5-5)
with T = JD - 2444239. 5 and i = xf^l.
Here m I M\j is the ratio of the mass of the satellite to the mass of Uranus; n is
the daily mean motion; L is the mean longitude measured from the ascending node
of Uranus' equator on the Earth's equator to the ascending node of the satellite
orbit on Uranus' equator along the orbit to the mean position of the satellite, z and
( are converted to classical elements with the following formulas
z = e exp((P),
7 (6.5-6)
C = sin — exp(((9),
where e is the eccentricity; P is the longitude of the pericenter reckoned from the
ascending node of Uranus' equator on the Earth's equator of 1950.0, along Uranus'
equator to the ascending node of the orbit on Uranus' equator, and then along the
orbit to the pericenter; 7 is the inclination of the orbital plane to Uranus' equator;
6 is the longitude of the ascending node of the orbit on Uranus' equator measured
from the ascending node of Uranus' equator on the Earth's equator.
Referred to the Earth's equator the motions of the satellites are direct. However,
referred to the ecliptic they are retrograde, having an angle of inclination of roughly
98°. The orbital plane is inclined at such a large angle to the Earth's equator
that the semi-major axis of the apparent orbit usually lies nearly north and south.
Consequently, the greatest elongations are designated as northern and southern
elongations instead of eastern or western. Only when the Earth is near the plane
through the celestial pole perpendicular to the orbital plane, as during 1986-1987,
is the minor axis of the apparent orbit directed approximately north and south.
Even then, the north pole of the orbital plane may lie to the south of the geocentric
position of Uranus. The north pole of the orbital plane is that from which the
satellites' motion appears to be counterclockwise. This pole corresponds to the
physical ephemeris south pole of Uranus, which is defined as lying south of the
invariable plane of the solar system.
During the course of one revolution of Uranus, the Earth passes twice through
the plane perpendicular to the orbital plane when the apparent orbits of the satel-
lites are almost circular; and likewise, it passes twice through the orbital plane when
the apparent orbits become straight lines. The Earth passed through the orbital
plane from south to north in 1966, and the sequence of geometric relations during
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 371
the interval from then until 2050 exemplifies the cycle during the 84-year period of
Uranus.
1966-2008: Earth north of orbital plane; B positive.
Superior conjunction at position angle P.
1966-1986: Northern elongation at P + 90°, when u - U = 90°.
1986: Earth passed through plane perpendicular to
orbital plane, at 17= 270°.
1986-2008: Northern elongation at P - 90°, when u - U = 270°.
2008: Earth passes through orbital plane, north to
south; B = 0°.
2008-2050: Earth south of orbital plane; B negative.
Superior conjunction at position angle P+ 180°.
2008-2030: Northern elongation at P - 90°, when u- U = 270°.
2030: Earth passes through plane perpendicular to
orbital plane, at If — 90°.
2030-2050: Northern elongation at P + 90°, when u - V = 90°.
At the passage of the Earth through the plane perpendicular to the orbital
plane, the position angle of the northern elongation changes by 180°, because the
more northerly and more southerly extremities of the major axis are interchanged.
The angle P is the position angle of the pole of the orbital plane that lies north of the
celestial equator; but either the eastern or western extremity of the minor axis may
be directed toward this pole, according to circumstances. Northern elongation may
be to either the east or west of north. The elongations cannot be unambiguously
designated as eastern and western.
To calculate the values of J, u, and N, consider spherical triangle ABD in Fig-
ure 6.5.1. Spherical trigonometry gives us:
sm.Jsin(N — Ne) = sin 7 sin 0,
(6.5-7)
sin/cos^ - Ne) = cos 7 sin /e +sin7COsJecos6l,
which can be solved for / and N. The values of Ne and Je, the longitude of the
ascending node of Uranus' equator and its inclination to the Earth's equator, are
found from the coordinates of Uranus' south pole: a0 = 187?43 and S0 = -15? 10
(Davies, 1989):
7Ve = a0 + 90°,
Je = 90° - So.
(6.5-8)
372
EXPLANATORY SUPPLEMENT
Figure 6.5.1
Reference system for the Uranian satellites. N = TD, P = AB + BC,
and L = AB + BC + mean anomaly.
The arc length DB is found from
sin DB sin 7 = $\n(N - Nc)s'mJe. (6.5-9)
Then
u = L-6 + DB + (f-M), (6.5-10)
where (/ — Af) is the equation of center which can be found from Equation 6.11-1,
and
M = L-P. (6.5-11)
Orbital, physical, and photometric data are presented on pages F2-F3 of The
Astronomical Almanac for 10 small satellites discovered by Voyager. The orbits of
these satellites all lie inside the orbit of Miranda. Cordelia and Ophelia are shepherd
satellites for the e-ring. It has also been proposed that Cordelia is the outer shepherd
for the <5-ring, and that Ophelia is the outer shepherd of the 7-ring.
g / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES
6 6 THE SATELLITES OF NEPTUNE
373
Ephemerides are given for the elongations and for the apparent distance and po-
sition angle of Triton, calculated from the orbital elements determined by Harris
(1984)- Differential right ascensions and declinations are given for Nereid from or-
bital elements computed by Mignard (1981).
6.61 Triton
The orbit of Triton is retrograde. The orbit, as far as it has been determined from
observations, is circular. The orbital plane is defined with respect to a fixed reference
plane. This reference plane is defined such that the projections of spin angular
momentum of Neptune and the orbital angular momentum of Triton on the plane
are equal and opposite. The pole of this fixed plane with respect to the Earth's
equator and equinox of J2000.0 are
ap = 298.° 72 + 2. 58 sin N - 0. 04 sin 2/V,
5P = 42.°63 - 1. 90cosJV + 0.01 cos 2JV,
(6.61-1)
N = 359.° 28 + 54.° 308 T,
T = (JD - 245 1 545. 0) / 36525.
The orbital elements of Triton referred to the fixed plane (see Figure 6.61.1) are
tQ = JD24332S2.5,
a = 488"49,
e = 0,
n = 61 "2588532/ day,
Lq = 200f 913,
7 = 158°996 = 21?004 retrograde,
9 = 151?40] +0?57806(JD - t0)/ 365. 25, (6.61-2)
where a is the greatest elongation at unit distance; e is the eccentricity; « is the
daily mean motion; Lq is the mean longitude reckoned from the ascending node
through the invariable plane at epoch; 7 is the inclination of the orbital plane to
the invariable plane; and 6 is the angle from the intersection of invariable plane with
the Earth's equatorial plane of 1950.0 to the ascending node of the orbit through
the invariable plane.
Because the retrograde direction of the motion is represented by an orbital
inclination greater than 90°, the ascending node of the orbit is the point at which
374
EXPLANATORY SUPPLEMENT
Figure 6.61.1
Reference system for Triton. N = TC.
the satellite crosses the equator from south to north. The pole of the orbit from
which the motion appears counterclockwise is the south pole, at position angle P.
Elongation is at position angle P - 90°, when u—U= 270°.
Twice during the course of one revolution of Neptune, about 165 years, the
Earth passes through the orbital plane of Triton, when the apparent orbit becomes
a straight line. The Earth crossed the orbital plane from north to south near the end
of 1952; for an interval during 1950-1954, Triton transited the disk of Neptune and
was occulted by the disk during each revolution. Before 1953, B was negative, and
inferior conjunction was on the southern arc of the apparent orbit at position angle
P. Since the passage through the orbital plane, B has been positive, and inferior
conjunction is on the northern arc at P + 180°. The Earth will pass through the
orbital plane again in 2035.
When the numerical value of B reaches a maximum as the Earth passes through
the plane perpendicular to the orbital plane, as in 1987, the minor axis of the
apparent orbit lies exactly north and south, the major axis lies east and west, and
the position angle of the more northerly elongation changes by 180°. Before 1987,
the western elongation was the more northerly; since then, the eastern elongation
has been the more northerly.
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 375
The computation of J, u, and N is as follows. From the spherical triangle in
Figure 6.61.1 we have
sin Jsin(Af ~ Nt) = sin i sin 8,
(6.61-3)
sin J cos(W — Ne) = cos i sin Je + sin i cos 7e cos 8,
which can be solved for J and N. Ns and 7e, the longitude of the node and the
inclination of the invariable plane to the Earth's equator, are found from
Ne = ap+ 90°,
Je = 90° - <5p.
(6.61-4)
The arc length CB can be computed from
sin CB sin i = sin(iV - Ne) sin 7e. (6.61-5)
u follows from
u=Lo+DB + n(JD - 10), (6.61-6)
where JD is the Julian date of interest.
6.62 Nereid
The orbit of Nereid is unique among the satellites in the solar system, because of its
extreme eccentricity and the great difference in its inclination from that of Triton.
The orbital elements of Nereid with respect to the orbital plane of Neptune are (see
Figure 6.62.1):
a = 0. 036868 AU,
e = 0. 74515 - 0. 006 cos 29 + 0. 0056 cos(2u> - 2(6),
cos 7 = cos 10?041 - 0?0094cos 29,
0 = 329?3 -2?4r+ 19?7sin2!?- 303sin4!?
+ 0°7sin6<? + 0?357 sin 20+ l°276sin(2u; - 24>),
P=9 - 19?25sin2^ + 3?23sin4!P
- 0°725 sin 69 - 0°351 sin 24> - 0°7 sin(2w - 2(6),
M = 358? 91 + nt - 0?38 sin2# + l?0sin(2w - 20),
n = 0?999552/day, (6.62-1)
where 9 ■ 282?9+2°68 T and 2<f> = 107?4+0f 01 196 f. T is reckoned in Julian centuries
and f in days from JD 2433680.5. The inclination, 7, is referred to Neptune's orbital
376
EXPLANATORY SUPPLEMENT
Figure 6.62.1
Reference system for Nereid. N = TD and P = AB + BC.
plane. The longitude of the node, 8, is measured in Neptune's orbital plane from
its intersection with the 1950.0 celestial equator. The orbit of Neptune is taken to
have inclination, Jc = 22?313, and longitude of ascending node, Nc = 3?522 with
respect to the 1950.0 celestial equator. M is the mean anomaly; u>, the argument of
pericenter; e, the eccentricity; and n, the mean daily motion.
The differential right ascension and declination x and y coordinates in the orbital
plane are computed from the Equations 6.15-3. These coordinates are transformed
to the orbital plane using
= R3(-0)R,(-7)R3(-P)
(6.62-2)
These coordinates are then transformed to the Earth equator and equinox of
1950.0 using
22 J
= R3(-JVe)R| (-./,)
(6.62-3)
g / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 377
The coordinates X2, y2, Z2 are then used to compute f and 77 with Equations
6.15-8-
6j THE SATELLITE OF PLUTO
Times of northern elongation, apparent distances, and position angles are presented
in The Astronomical Almanac for Charon. The orbital elements from which these
are calculated are taken from Tholen (1985). The elements, referred to the Earth's
equator and equinox of 1950.0, are
a m 0:'00012788,
n = 56?3625 / day,
J = 94?3,
/V = 223°7,
k = 78°6,
to = JD2445000. 5, (6.7-1)
where a is the semi-major axis; n, the mean daily motion; J, the inclination of the
orbital plane to the Earth equator of 1950.0; N, the longitude of the ascending node
reckoned from the vernal equinox; Iq, the mean longitude at epoch reckoned from
the ascending node along the orbit; and to, the date of epoch.
The orbit of Charon is retrograde. The pole of the orbit from which the motion
appears counterclockwise is the south pole, at position angle P. Elongation is at
position angle P - 90°, when u-U = 270°.
Twice during the course of one revolution of Pluto, about 248 years, the Earth
passes through the orbital plane of Charon, when the apparent orbit becomes a
straight line. The Earth crossed the orbital plane three times during 1987-1988,
the end result being a crossing from north to south. In the interval 1985-1990,
Charon transits the disk of Pluto and is occulted by the disk during each revolution.
Before 1987, B was negative, and inferior conjunction was on the southern arc of
the apparent orbit at position angle P. Since the passage through the orbital plane,
B has been positive, and inferior conjunction on the northern arc at P+ 180°.
When the numerical value of B reaches a maximum as the Earth passes through
the plane perpendicular to the orbital plane, as in 2050, the minor axis of the
apparent orbit lies exactly north and south, the major axis lies east and west, and
*e position angle of the more northerly elongation changes by 180°. Before 2050,
the western elongation is the more northerly; after 2050, the eastern elongation will
be the more northerly.
378 EXPLANATORY SUPPLEMENT
Since the orbit is referred to the Earth equator and equinox of 1950.0, / and N
are already known, u is found simply from
u = l0+ fi(JD - t0), (6.7-2)
where JD is the Julian date of interest.
6.8 REFERENCES
Adams, J.C. (1877). "Continuation of Tables I and III of Damoiseau's Tables of Jupiter's
Satellites" Nautical Almanac for 1881 Appendix, 15-23.
Aksnes, K. (1978). "The Motion of Jupiter XIII (Leda), 1974-2000" Astron. J. 83, 1249-
1256.
Andoyer, H. (1915). "Sur le Calcul des Ephemerides des Quatres Anciens Satellites de
Jupiter" Bull. Astr. 32, 177-224.
Arlot, J.-E. (1982). "New Constants for Sampson-Lieske Theory of the Galilean Satellites
of Jupiter" Astron. Astrophys. 107, 305-310.
Bessel, F.W. (1875a). "Untersuchungen fiber den Planeten Saturn, seinen Ring und scinen
vierten Trabanten" Abhandlungen I, 110-127.
Bessel, F.W. (1875b). "Bestimmung der Lage und Grosse des Saturnsringes und der Fugur
und Grosse des Saturns" Abhandlungen I, 150-159.
Bessel, F.W. (1875c). "Ueber die Neigung der Ebene des Saturnsringes" Abhandlungen I,
319-321.
van Biesbroeck, G. (1957). "The Mass of Neptune from a New Orbit of Its Second Satellite
Nereid" Astron. J. 62, 272-274.
Bobone, J. (1937a). "Tablas del VI (sexto) Satelite de Jupiter" Ast. Nach. 262, 321-346.
Bobone, J. (1937b). "Tablas del VII (septimo) Satelite de Jupiter" Ast. Nach. 263, 401-
412.
Born, G.H. and Duxbury, T.C. (1975). "The Motions of Phobos and Deimos from Mariner
9 TV Data" Cel. Mech. 12, 77-88.
Brouwer, D. and Clemence, G.M. (1961). Methods of Celestial Mechanics (Academic Press,
London), chapter XVI.
Christy, J.W. and Harrington, R.J. (1978). "The Satellite of Pluto" Astron. J. 83, 1005-
1008.
Cohn, F. (1897). "Bestimmung der Bahnelemente des V. Jupitersmondes" Ast. Nach. 142,
289-338.
Comrie, L.J. (1934). "Phenomena of Saturn's Satellites" Mem. B.A.A. 30, 97-106.
Crawford, R.T. (1938). "The Tenth and Eleventh Satellites of Jupiter" P.A.S.P. 50, 344-
347.
de Damoiseau, M.C.T. (1836). "Tables Ecliptiques des Satellites de Jupiter" Paris.
Davies, M.E., Abalakin, V.K., Bursa, M., Hunt, G.E., Lieske, J.H., Morando, B., Rapp,
R.H., Seidelmann, P.K., Sinclair, A.T., and Tjuflin, Y.S. (1989). "Report of the
IAU/IAG/Cospar Working Group on Cartographic Coordinates and Rotational Ele-
ments of the Planets and Satellites: 1988" Cel. Mech. 46, 187-204.
Delambre, J.B.J. (1817). "Tables Ecliptiques des Satellites de Jupiter" Paris.
Dunham, D.W. (1971). "The Motions of the Satellites of Uranus" Dissertation. (Yale
University, New Haven, CT).
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 379
Eichelberger, W.S. (1892). "The Orbit of Hyperion" Astron. J. H, 145-157.
Eicbelberger, W.S. and Newton, A. (1926). "The Orbit of Neptune's Satellite and the Pole
of Neptune's Equator" A.P.A.E. 9, 275-337.
Elliot, J.L., French, R.G., Frogel, J.A., Elias, J.H., Mink, D.J., and Liller, W. (1981).
"Orbits of Nine Uranian Rings" Astron. J. 86, 444-455.
gsposito, L.W., Cuzzi, J.N., Holberg, J.H., Marouf, E.A., Tyler, G.L., and Porco, C.C.
(1984). "Saturn's Rings: Structure, Dynamics, and Particle Properties" in Saturn T.
Gerhels and M.S. Matthews, eds. (University of Arizona Press, Tucson, AZ).
Garcia, H.A. (1972). "The Mass and Figure of Saturn by Photographic Astrometry of Its
Satellites" Astron. J. 77, 684-691.
Grosch, H.R.J. (1948). "The Orbit of the Eighth Satellite of Jupiter" Astron. J. 53, 180-
187.
Hall, A. (1878). "Observations and Orbits of the Satellites of Mars" Washington Obser-
vations for 1875 Appendix I, 1-46.
Hall, A. (1885). "The Orbit of Iapetus, the Outer Satellite of Saturn" Washington Obser-
vations for 1882 Appendix I, 1-82.
Hall, A. (1886). "The Six Inner Satellites of Saturn" Washington Observations for 1883
Appendix I, 1-74.
Hall, A. (1898). "The Orbit of the Satellite of Neptune" Astron. J. 19, 65-66.
Harper, D., Taylor, D.B., Sinclair, A.T., and Kaixian, S. (1988). "The Theory of the
Motion of Iapetus" Astron. Astrophys. 191, 381-384.
Harrington, R.J. and Christy, J.W. (1981). "The Satellite of Pluto III" Astron. J. 86,
442-443.
Harris, AW. (1984). "Physical Properties of Neptune and Triton Inferred from the Orbit
of Triton" NASA CP-2330, 357-373.
Harris, D.L. Unpublished work on the satellites of Uranus, quoted by Kuiper (1956, p.
1632).
Harshman, W.S. (1894). "The orbit of Deimos" Astron. J. 14, 145-148. Harshman's el-
ements of Phobos "communicated by Prof. Harkness" are attached (in manuscript)
to a copy of this paper. It is believed that they were never published. See also Publ.
U.S.N.O. 6 Appendix I, B21 (1911).
Herget, P. (1938). "Jupiter XI" Harvard Announcement Card. J\63.
Herget, P. (1968). "Outer Satellites of Jupiter" Astron. J. 73, 737-742.
Herrick, S. (1952). "Jupiter IX and Jupiter XIP P.A.S.P. 64, 237-241.
Kendall, E.O. (1877). An Extension for the Year 1880 of Damoiseau's Tables of the Satel-
lites of Jupiter (probably never published).
Kozai, Y. (1957). "On the Astronomical Constants of Saturnian Satellites System" Ann.
Tokyo Obs. Ser. 2 5, 73-106.
Kuiper, G.P. (1949). "The Fifth Satellite of Uranus" P.A.S.P. 61, 129.
Kuiper, G.P. (1956). "On the Origin of the Satellites and the Trojans" Vistas in Astronomy
2, 1631-1666.
de Lalande, J.J. (1792). A stronomie 3rd edition, Paris.
Laskar, J. and Jacobson, R.A. (1987). "An Analytical Ephemeris of the Uranian Satellites"
Astron. Astrophys. 188, 212-224.
Levin, A.E. (1931, 1934). "Mutual Eclipses and Occultations of Jupiter's Satellites" J. B.A.J
42, 6-14 (1931). Mem. B.A.A. 30, 149-183 (1934).
Lieske, J.H. (1978). "Theory of Motion of Jupiter's Galilean Satellites" Astron. Astrophys.
56, 333-352.
380 EXPLANATORY SUPPLEMENT
Lieske, J.H. (1987). "Galilean Satellite Evolution: Observational Evidence for Secular
Changes in Mean Motions" Astron. Astrophys. 176, 146-158.
Marth, A. Marth published a long series of ephemerides for physical observations of the
planets and for observations of the satellites. The first and last of each series are alone
quoted here.
Marth, A. (1870-1894). "Ephemeris of the Satellites of Uranus" Mon. Not. Roy. Astron
Soc. 30, 150 154, 171.
Marth, A. (1873-1895). "Ephemeris of the Satellites of Saturn" Mon. Not. Roy. Astron
Soc. 33, 513-555, 164.
Marth, A. (1878-1891). "Ephemeris of the Satellites of Neptune" Mon. Not. Roy. Astron
Soc. 38, 475-451, 563.
Marth, A. (1883-1896). "Data for Finding the Positions of the Satellites of Mars" Mon.
Not. Roy. Astron. Soc. 44, 29-56, 435.
Marth, A. (1891-1896). "Ephemeris for Computing the Positions of the Satellites of
Jupiter" Mon. Not. Roy. Astron. Soc. 51, 505-556, 434.
Mignard, F. (1981). "The Mean Elements of Nereid" Astron. J. 86, 1728-1729.
Newcomb, S. (1875). "The Uranian and Neptunian Systems" Washington Observations
for 1873 Appendix I, 1-74.
Nicholson, S.B. (1944). "Orbit of the Ninth Satellite of Jupiter" Astrophys. J. 100, 57-62.
Pottier, L. (1896). "Addition aux Tables Ecliptiques des Satellites de Jupiter de Damoi-
seau" Bull. Astr. 13, 67-79 and 107-112.
Robertson, J. (1924). "Orbit of the Fifth Satellite of Jupiter" Astron. J. 35, 190-193.
Rohde, J.R. (1990). Unpublished.
Rose, L.E. (1974). "Orbit of Nereid and the Mass of Neptune" Astron. J. 79, 489-490.
Ross, F.E. (1905). "Investigations on the Orbit of Phoebe" Harvard Annals 53, 101-142.
Ross, F.E. (1907a). "Semi-definitive Elements of Jupiter's Sixth Satellite" Lick Obs. Bull.
4, 110-112.
Ross, F.E. (1907b). "New Elements of Jupiter's Seventh Satellite" Ast. Nach. 174, 359-
362.
Sampson, R.A. (1910). Tables of the Four Great Satellites of Jupiter (William Wesley
Son, London).
Sampson, R.A. (1921). "Theory of the Four Great Satellites of Jupiter" Mem. Roy. Astron.
Soc. 63, 1-270.
Shampine, L.F. and Gordon, M.K. (1975). Computer Solutions of Ordinary Differential
Equations (W.H. Freeman, San Francisco, CA).
Sinclair, A.T. (1974). "A Theory of the Motion of Iapetus" Mon. Not. Roy. Astr. Soc.
169, 591-605.
Sinclair, A.T. (1977). "The Orbits of Tethys, Dione, Rhea, Titan and Iapetus" Mon. Not.
Roy. Astr. Soc. 180, 447-459.
Sinclair, A.T. (1978). "The Orbits of the Satellites of Mars" Vistas in Astron. 22, 133-140.
Sinclair, A.T. (1989). "The Orbits of the Satellites of Mars Determined from Earth-based
and Spacecraft Observations" Astron. Astrophys. 220, 321-328.
Struve, G. "Neue Untersuchungen im Saturnsystem" Veroffentlichungen der Universitats-
sternwarte zu Berlin- Babelsberg 6, Parts 1, 4, 5. (1924) Part 1 Die Bahn von Rhea
1-16. (1930) Part 4 Die Systeme Mimas-Tethys und Enceladus- Dione 1-61. (1933)
Part 5 Die Beobachtungen der ausseren Trabanten und die Bahnen von Titan und
Japetus 1-44.
Struve, H. (1888). "Beobachtungen der Saturnstrabanten" Supplement I aux Observations
de Poulkova 1-132.
6 / ORBITAL EPHEMERIDES AND RINGS OF SATELLITES 381
Struve, H. (1898). "Beobachtungen der Saturnstrabanten" Publications de VObservatoire
Central Nicolas, Serie II II, 1-337.
Struve, H. (1903). "Neue Bestimmung der Libration Mimas-Tethys" Ast. Nach. 162, 325-
344.
Struve, H. (1911). "Uber die Lage der Marsachse und die Konstanten im Marssystem"
Sitzungsberichte der Koniglich Preussischen Askademie der Wissenschaften fur 1911
1056-1083.
Struve, H. (1913). "Bahnen der Uranustrabanten. Abteilung I: Oberon und Titania" Ab-
handlungen der Konigich Preussischen Akademie der Wissenschaften
Synnott, S.P. (1984). "Orbits of the Small Inner Satellites of Jupiter" Icarus 58, 178-181.
Taylor, D.B. (1984). "A Comparison of the Theory of the Motion of Hyperion with Ob-
servations Made During 1967-1982" Astron. Astrophys. 141, 151-158.
Taylor, D.B. and Shen, K.X. (1988). "Analysis of Astrometric Observations from 1967 to
1983 of the Major Satellites of Saturn" Astron. Astrophys. 200, 269-278.
Taylor, S.W. (1951). "On the Shadow of Saturn on Its Rings" Astron. J. 55, 229 230.
Tholen, D.J. (1985). "The Orbit of Pluto's Satellite" Astron. J. 90, 2353-2359.
Todd D.P. (1876). "A Continuation of de Damoiseau's Tables of the Satellites of Jupiter
to the Year 1900" (Washington Bureau of Navigation, Washington).
Wargentin, P.W. (1746). "Tabulae pro calculandis eclipsibus satellitum Jovis" in Acta Soc.
reg. sci. Upsaliensis ad annum 1747 (Laurentii Salvii, Stockholm).
Wilson, R.H., Jr. (1939). "Revised Orbit and Ephemeris for Jupiter X" P. A. S.P. 51, 241-
242.
van Woerkom, A.J.J. (1950). "The Motion of Jupiter's Fifth Satellite, 1892-1949" A.P.A.E.
13, 1-77.
Woltjer, J. (1928). "The Motion of Hyperion" Annalen van de Sterrewacht te Leiden 16,
part 3, 1-139.
Woolhouse, W.S.B. (1833). "New Tables for Computing the Occultations of Jupiter's
Satellites by Jupiter, the Transits of the Satellites and Their Shadows Over the Disc
of the Planet, and the Positions of the Satellites with Respect to Jupiter at Any
Time" Nautical Almanac for 1835 Appendix, 1-39.
Zadunaisky, RE. (1954). "A Determination of New Elements of the Orbit of Phoebe,
Ninth Satellite of Saturn" Astron. J. 59, 1-6.
CHAPTER
Physical Ephemerides of the Sun, Moon,
Planets, and Satellites
by James L. Hilton
7.1 INTRODUCTION
The physical ephemerides of the Sun, Moon, planets, and satellites give information
on the apparent physical aspects of the disks of those objects. The physical ephem-
erides are used for making and reducing observations of the surface markings of an
object and for determining the exact center of an object's disk from observations
of the limbs of the object. All the tabulated values are geocentric and have been
corrected for the effects of aberration. Hence the information tabulated gives the
appearance of the object as seen from the geocenter at the tabulated time.
In general, the ephemerides can be broken into two groups: (1) information
depending on the disk of the object alone, the rotational elements, and cartographic
coordinates, and (2) the phase and magnitude of the object, which depend on the
position of the Earth and Sun relative to the object and the object's physical
appearance.
Unless otherwise stated, all coordinates given in this chapter refer to the mean
equator and equinox of J2000.0. T is the time in Julian centuries of 36525 days from
J2000.0, and t is the time in days (86400 SI seconds) from J2000.0. The J2000.0
coordinate system is defined by the FK5 star catalog and has the standard epoch
of 2000 January 1.5 (JD 2452545.0), TDB.
7.1 1 Rotational Elements and Cartographic Coordinates
The cartographic coordinates of an object consist of the orientation of the north
Pole and the position of the prime meridian of the object with respect to an inertial
reference frame. These elements are given by the three quantities: ao, 6q, an<J W ^
a function of time. These parameters are shown in Figure 7.11.1.
383
384
EXPLANATORY SUPPLEMENT
Figure 7.11.1
The position of the north pole and prime meridian
of a planet in Earth equatorial coordinates
The quantities qq and <5o are the right ascension and declination of date of the
object's north pole. The IAU definition of the north pole of a planet, from The IAU
Working Group on Cartographic Coordinates and Rotational Elements of Planets
and Satellites (Davies et al., 1991), is the rotation axis that lies on the north side
of the invariable plane of the solar system, an and <5n may vary slowly with time
because of the precession of the object about its polar axis. In the absence of other
information, the axis of rotation is assumed to be normal to the mean orbital plane.
This assumption is used for Mercury and most of the satellites.
The angle W is measured along the object's equator, in a counterclockwise
direction when viewed from the north pole, from the ascending node of the object's
equator on the Earth's mean equator of J2000.0 to the point where the prime
meridian of the object crosses its equator. The right ascension of the node (£?) is at
6h+an, and the inclination of the object's equator to the standard equator is 90° — bo-
Generally, the prime meridian is assumed to rotate uniformly with the object, hence
W varies linearly with time. For objects without observable, fixed surface features
the adopted expression for W defines the prime meridian and is not subject to
correction. However, if a cartographic position for the prime meridian is assigned
to an object, that is, if the position of the prime meridian is defined by a suitable
observable surface feature, the expression for W is chosen so that the ephetnens
position follows the motion of the cartographic position. Although the definition of
7 i PHYSICAL EPHEMERIDES OF THE SUN, MOON. PLANETS. AND SATELLITES 335
the ephemeris prime meridian is chosen to follow the cartographic prime meridian
as closely as possible, there may be errors in the rotational elements of the object
that will cause the cartographic position to drift away from the ephemeris position
by a small amount AW. The angle AW is measured positively counterclockwise
from the ephemeris position of W as viewed from above the north pole. For a prime
meridian that has a cartographic definition, the ephemeris definition of W may be
changed as more accurate information on the motion of the cartographic prime
meridian is obtained.
The definition of the north pole requires the rotation about the pole to be
classified as either direct or retrograde. Direct rotation is counterclockwise as viewed
from above the planet's north pole, while retrograde rotation is clockwise. For a
planet with direct rotation the angle W increases with time. For most of the satellites
it is assumed that the rotation period is equal to the mean orbital period.
Information on the point defining the prime meridian and pole for the Sun can
be found in Carrington (1863) and Section 7.2 of this supplement. The definition
for the prime meridian and pole of each planet can be found in Davies et al. (1991)
and Sections 7.41-7.48 of this supplement. Information defining the prime meridian
and pole of the primary planetary satellites can be found in Davies et al. (1991) and
Sections 7.3 and 7.51 through 7.56 of this supplement. In general, the expressions
defining aoi <5oi and W should be accurate to one tenth of a degree. Two decimal
places are given, however, to assure consistency when changing coordinate systems.
Zeros are added to the rate values (W) for computational consistency and are not
an indication of significant accuracy. Three significant digits beyond the decimal
point are given in the expressions for the Moon and Mars, reflecting the greater
confidence in their accuracy. The recommended coordinate system for the Moon
is the mean Earth-polar axis system in contrast to the principal axis system. The
Earth-polar axis system uses the pole of rotation of the Moon and the mean axis of
the Moon, which points toward the center of the Earth. The principal axis system
uses the axes of the rotation and the defined 0° meridian of the object.
Both planetocentric and planetographic systems of coordinates are used in the
study of the planets and satellites. Both systems are based on the same fundamen-
tal axis of rotation but differ, as explained below, in their definitions of latitude,
longitude, and range. Planetocentric coordinates are used for general purposes and
are based on a right-handed system of axes with its origin at the center of mass of
the object. Planetographic coordinates are used for cartographic purposes and de-
pend on the adoption of a reference surface, usually a spheroid, that approximates
an equipotential surface of the object. The latitude and ranges for both of these
coordinate systems are shown in Figure 7.11.2.
The reference surfaces for most of the planets are spheroids for which the equa-
torial radius, a, is larger than the polar radius, b (see Figure 7.11.2). For some of
tne planets and most of the satellites the reference surface is a sphere. For some
satellites, such as Phobos, Deimos, and Hyperion, the reference surface should be
EXPLANATORY SUPPLEMENT
Planet's
axis of rotation
Figure 7.11.2
Planetocentric and planetographic coordinate systems. The quantities shown are: 4>, the
planetocentric latitude; r, the distance from the center of mass; <f>' the planetographic latitude; and
h, the height of a point above or below [he planetographic reference surface. Also included are o,
the equatorial radius of the object; and b, the polar radius.
triaxial. Triaxial ellipsoids would make many computations more complicated, espe-
cially those related to map projections. Therefore, spherical or spheroidal reference
surfaces are usually used in mapping programs. The polar axis for each reference
surface is assumed to be the mean axis of rotation as defined by the adopted ro-
tational elements since the present accuracy of measurement does not reveal a
deviation between the axis of rotation and the axis of figure for most of the planets.
The radii and flattening of the planets are found in The Astronomical Almanac and
in Section 7.4 of this supplement. The mean radii of the natural satellites of the
planets are also found on page F3 of The Astronomical Almanac.
Many small bodies in the solar system (satellites, asteroids, and comet nu-
clei) have very irregular shapes. Sometimes spherical reference surfaces are used
to preserve projection properties. Orthographic projections are often adopted for
cartographic purposes because they preserve the irregular appearance of the body
without artificial distortion.
In both systems, the fundamental reference Z-axis is the mean axis of rotation
with the positive direction being toward the north pole. The equator is the plane,
passing through the center of mass of the planet, perpendicular to the z-axis. The
JC-axis is defined by the intersection of the plane of the equator and the plane con-
taining the prime meridian. The choice of the prime meridian is arbitrary. The
7 / PHYSICAL EPHEMERIDES OF THE SUN. MOON, PLANETS. AND SATELLITES 387
y-axis is then orthogonal to the other two axes and makes a right-handed coordi-
nate system. The angular coordinates used in these two systems are latitude and
longitude. The range of the planetocentric coordinate system is the distance from
the center of mass, and the range of the planetographic system is the height above
0r below the reference surface.
Latitude (see Figure 7.11.2) is measured north and south of the equator with
the northern latitudes being designated as positive and the southern latitudes as
negative. The planetocentric latitude of a point on the surface of a planet is the
angle <j> between the equatorial plane and the line segment connecting the point to
the center of mass. The planetographic latitude of a point is the angle 4>' between
the line segment normal to the equipotential surface from the point in question to
the equatorial plane. Since the equipotential surface is usually a spheroid, the shape
of the spheroid is related to that of a sphere by its flattening, /,
f = (a-b)ia, (7.11-1)
where a is the equatorial radius of the object and b is the polar radius of the object.
For a spherical object, the planetocentric and planetographic latitudes are the same.
The values for the radii of the planets and axes given in Sections 7.41-7.48
and in The Astronomical Almanac are derived by different methods, and do not
always refer to common definitions. Some radii use star or spacecraft occultation
measurement, some use limb fitting, and some use control-network computations.
For example, the spheroid for the Earth refers to mean sea level, a definition that
can be used only for this planet. The radii and axes of the large gaseous planets,
Jupiter, Saturn, Uranus, and Neptune refer to a one-bar pressure surface. The radii
in the tables are not necessarily the appropriate values to be used in dynamical
studies; the radii actually used to derive a value of J%, for example, should always
be used in conjunction with that value.
Longitude is measured along the equator from the prime meridian. Planetocen-
tric longitudes, A, are measured positively in the counterclockwise direction when
viewed from the north pole. Planetographic longitudes, A', are measured positively
in the direction opposite to the rotation of the object. For an object with a direct
rotation, the planetographic longitude increases in the clockwise direction when
viewed from above the north pole. Planetocentric longitudes are measured from
the ephemeris position of the prime meridian as defined by the rotational elements,
but the planetographic longitudes are measured from the cartographic position of
the prime meridian as defined by the adopted longitude of some clearly observable
surface feature. For the Earth, the Sun, and the Moon, longitudes are measured
from 0° to 180° east and west, and east longitude (counterclockwise when viewed
from above the north pole) is commonly considered to be positive.
388 EXPLANATORY SUPPLEMENT
In the planetographic system, a point P not on the reference surface is specified
by the planetographic longitude and latitude of the point P' on the reference sur-
face at which the normal line segment connecting P to the equatorial plane passes
through the reference surface and the height of P above or below the reference
surface.
7.12 Phases and Magnitudes
The tabulated magnitude of an object is the integrated total visual magnitude.
The magnitudes for the planets are based on Harris (1961). The magnitude of a
planet is given in the section on that planet in this chapter. The magnitudes for
the brightest asteroids are taken from Lumme and Bowell (1981). The basis for the
photometric radii of the asteroids is based on Lumme, Karttunen, Piironen, and
Bowell (1986) and is described in Section 7.6. Rotation rates for the asteroids are
not included in The Astronomical Almanac, but information on how rotation curves
are determined can be found in Lumme, Karttunen, Bowell, and Poutanen (1986),
The apparent magnitude of an object is a function of its distance from the Sun,
r, its distance from the Earth, d, and the solar phase angle, / (see Figure 7.12.1).
The solar phase angle is the angular distance at the object between the Sun and
the Earth. For Mercury, Venus, and the Moon, / varies from 0° to 180°. For Mars
and its satellites / can be as large as 47°. For the rest of the planets, satellites, and
most of the asteroids i reaches only a few degrees.
Neglecting variations from rotation or other intrinsic causes, the observed mag-
nitude for an object is given by
V = V(1,0) + 5 log,0(rf) + Am(i), (7.12-1)
where V( 1 , 0) is the magnitude of the planet seen at a distance of 1 AU with a phase
angle of 0°, d is the Earth-planet distance in AUs, and Am(i) is the correction for
the variation in brilliancy of the planet with phase angle.
The quantity Am(i) is measured empirically. It is the result of two effects. The
first source of variation arises from the fraction of the illuminated disk visible from
the Earth as it varies with /'. The second effect is due to the properties of diffuse
reflection from the object's surface and/or atmosphere. The parameter Am(0 's
expressed as a power series in i, retaining as many terms as necessary to represent
the observations. For the outer planets a single term is usually sufficient, hence
Am(i) is usually referred to as the phase coefficient. The series or coefficients for
each planet and the Galilean satellites of Jupiter are presented in Sections 7.4 and
7.5.
If the intrinsic properties of a planet cause its brightness to vary, the problem
of predicting the apparent magnitude at a given time is considerably more compli-
cated. The causes of intrinsic variability are (1) the existence of surface markings on
PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 389
Kasic wctots and the angle of illumination. These quantities affect the illumination
of a planet as seen from the Earth. The figure is drawn in the Earth-Sun-planet plane.
the planet, and (2) changes in the planet's atmosphere or volatile surface markings
such as Mars' ice caps. The surface markings on an object give rise to brightness
variations that depend on the rotation of the object. The second source of intrinsic
variability gives rise to variations of an irregular nature. The atmospheric varia-
tions may be short-lived, such as sand storms on Mars, or persist for long periods
of time, such as the Great Red Spot on Jupiter. Because the atmospheric variations
change erratically, their effects on a planet's albedo cannot be predicted with great
accuracy although some, such as Mars' ice caps, generally follow a regular cycle^
For some purposes it is convenient to use the mean opposition magnitude, V0,
which is related to V(1,0) by
V() = V(l,0) + 51og10[fl(fl- 1)1 (7.12-2)
where a is the semi-major axis of the planet's orbit in astronomical units.
390 EXPLANATORY SUPPLEMENT
The surface brightness (SB) of an object is the average brightness value for the
illuminated portion of the apparent disk. The units for the surface brightness are
visual magnitudes per square arcsecond. This is represented mathematically by
SB =V +2.5 logl0(k7iab'), (7. 12-3)
where a is the equatorial radius of the object, b' is the apparent polar radius of
the object that will be given in equation 7.12-6, and k is the fraction of the object
illuminated that will be given in Equation 7.12-32.
The apparent disk of an object is always an ellipse. The apparent flattening
is always less than or equal to the flattening of the object itself (depending on
the apparent tilt of the objects axis). To derive the apparent polar radius and the
apparent flattening of an object, start with the equation for an ellipse. The ellipse
can be represented parametric ally by
r={x2+y1)m,
* = acos7, (7.12-4)
y = b sin 7,
where x is the length of the projection of the point along the major axis from the
center of the ellipse, y is the length of the projection of the point along the minor
axis from the center, a is the length of the major axis, b is the length of the minor
axis, and 7 is the angle between the line segment connecting the point with the
center of the ellipse and the semi-major axis (see Figure 7.12.2). For a planet, the
apparent polar radius is the length of the radius, R, whose planetocentric latitude
is the same as the planetocentric colatitude of the sub-Earth point, and 7 is the
latitude of the sub-Earth point. Using the definitions for x and y in the equation
for R
R = (a2 cos2 7 + b2 sin2 7)1/2. (7.12-5)
Rearranging the definition for flattening, Equation 7.11-1, gives
b = a(\ -/).
(7.12-6)
So
R - (a2 cos2 7 + a2(l - f)2 sin2 T)1/2
= a(cos2 7 + sin2 7-2/ sin2 7 + f sin2 7) 1/2
= a(l - 2/ sin2 7 + f sin2 7) 1/2 . (7.12-7)
PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES
391
Figure 7.12.2
The geometric appearance of the apparent polar radius, b'
Using the binomial expansion:
y = fl[l - \(2f sin2 j-fi sin2 7) + 5(2/ sin2 7 - f sin2 1)2 - ■■■]. (7.12-8)
2 o
However, for the planets / <C 1, so terms of order f or larger will be negligible,
and
R = a{ 1 - / sin2 7)
= a(l - /cos2 8)
= fl{l -/[(// -b)/(|f/||b|)]2},
(7.12-9)
where b' is the vector from the planetocenter to the apparent polar axis, b is the
vector from the planetocenter to the true polar axis, and 9 is the planetocentric
colatitude of b' . The apparent flattening of the planet will then be
f = (a-b')la.
(7.12-10)
The remaining tabulated quantities are most easily understood from Figure 7.12.3,
which shows the apparent disk of an object as it is seen from the Earth. At the cen-
ter of the apparent disk is the sub-Earth point, e. Other points of reference are the
392
North
EXPLANATORY SUPPLEMENT
South
Figure 7.12.3
The disk of a planet as seen by an observer on the Earth. The quantities shown are: e, the sub-Earth
point; s, the subsolar point; ds the apparent distance from the sub-Earth point to the subsolar point;
PA, the position angle of the subsolar point; n, the north pole of the planet; dn the apparent distance
from the north pole from the sub-Earth point; PA„, the position of the north pole; and Q, the
geometric maximum defect of illumination.
subsolar point, s, and the north pole, n. For most of the planets, the planetographic
longitude and latitude of the points e and s are tabulated. The apparent positions
of the subsolar point, s, and the north pole, n, with respect to the apparent center
of the disk (point e) are tabulated by the apparent distances, dn and ds, and posi-
tion angles, PA„ and PAS for n and s, respectively. The position angle is measured
eastward from north, N, on the celestial sphere. The direction north being denned
by the great circle on the celestial sphere passing through the center of the object's
apparent disk and the true celestial north pole of date. The apparent distance of
a tabulated point is listed as positive if the point lies on the visible hemisphere of
the object and negative if the point is on the far side of the object. Thus, as point
n or j passes from the visible hemisphere to the far side, or vice versa, the sign of
the distance changes abruptly, but the position angle varies continuously. However,
when the point passes close to the center of the disk, the sign of the distance re-
mains unchanged, but both the distance and position angle may vary rapidly and
may appear to be discontinuous in the fixed interval tabulations.
The planetographic coordinates of the sub-Earth and subsolar points are calcu-
lated in the following manner. First, the planetary ephemerides are used to calculate
? j PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 393
the heliocentric ecliptic rectangular coordinates for the planet, r = (x,y,z), and the
Earth, re = (*e,ye,Ze) at a given Terrestrial Julian Date (TJD). The vector from the
Earth to the object is then determined by subtracting re from r;
d = r - re
= (x-xe,y-ye,z- Ze)
= (*a,ya,3i). (7.12-11)
Next, the light-time correction is found by determining the length of d and dividing
by the speed of light, c, to obtain the time:
|d| = (d-d)"2 = (^+^ + ^)1/2>
f=|d|/c. (7.12-12)
The difference TJD — t is calculated to approximate the time at which the image
of the planet visible at the Earth at TJD left the planet. This time, TJD — t, is used
to determine the approximate position of the planet, T\ - (x\,y\,zi), as seen from
the Earth at time TJD. If desired, the previous process can be iterated to produce
a more accurate planetary position; however, planetary motions are slow enough
that to the accuracy of The Astronomical Almanac only one iteration is required.
The position of the Earth, re, and the apparent position of the planet, t\t are
then transformed from the mean equator and ecliptic of J2000.0 to the equator
and ecliptic of date using the precession routines described in Section 3.21 and the
nutation routines described in Section 3.22. The precessed and nutated position
vectors are designated red for the Earth and rd for the planet. A new vector d is
computed by subtracting rd from red as in Equation 7.12-11 and then correcting for
aberration in position given in Section 3.25. The unit vector along the planet-Sun
line, j, is a vector of unit length denoting the direction of the Sun from the planet.
The vector j is found by dividing -rd by its length, |rd|. The unit vector along the
planet-Earth line, jd, is found in a similar manner.
The position of the north pole of the planet, (a0, <5n), is then found from Davies
et al. (1991). These coordinates are then precessed and nutated to the ecliptic and
equinox of date, (ad, 6d). A unit vector, n, in the ecliptic rectangular coordinates
is then calculated:
n = (cos6dcosad,cos<5dsinad,sini5d). (7.12-13)
This vector gives the planetocentric direction of the north pole of the planet in the
coordinates of the equator and ecliptic of date.
Finally, the ecliptic rectangular coordinates of the unit vector pointing from the
Planet's center to the intersection of the planet's prime meridian and the equator,
394
EXPLANATORY SUPPLEMENT
Figure 7.12.4
Planetocentric unit vectors for points of interest on the disk of the planet, and the angles between the
Earth's equator of J2000.0 and the planet's equator of date. The quantities depicted are: n, the north
pole of the planet; n, the north pole of the Earth's equator and equinox of J2000.0; fi, the
planetocentric unit vector for the north pole; s, the subsolar point; j, the planetocentric unit vector for
the subsolar point; ja, the planetocentric unit vector for the sub-Earth point; U, the planetocentric unit
vector pointing to the intersection of the object's prime meridian and its equator; W, the angle along
the object's equator from the ascending node on the Earth's equator of date to the object's prime
meridian; g, the angle along the great circle from the Earth's equator of J2000.0 to the Earth's north
pole of J2000.0, which goes from the Earth's equator to the intersection of the planet's prime meridian
and its equator of date; and, S, the angle along the Earth's equator from the ascending node of the
planet's equator to the point where the angle g reaches the equator. The planetocentric unit vector for
the sub-Earth point, jd points directly at the observer.
r , is needed. To obtain this vector, first determine the intermediate angles g and S
(Figure 7.12.4):
sing-sinWcos^, (7.12-14)
sin S = sin W sin Sd I cos g, (7,12-15)
cos S = cos W ! cos g. (7.12-16)
7 / PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 395
Since g runs from +90° to —90° (see Figure 7.12.3), there is no loss of generality in
not directly determining cosg. The right ascension and decimation of U are then
ctu = ad + 90° + tan_1(sinS/cosS), (7.12-17)
6V = tan" 1 (sing /cos g). (7.12-18)
The coordinates for the sub-Earth and subsolar points are then computed by using
the vectors for the north pole of the planet and U to determine the rotation from
the equatorial ecliptic coordinate system to give the planetocentric latitude and
the planetographic longitude. (The planetographic longitude is supplied from the
definition of W, and the planetocentric latitude is determined because the angles are
measured with respect to the axes of the coordinate system at the center of mass
of the planet and not the local normal on the equipotential surface.) The flattening
of the planet, f, is used to convert the planetocentric latitude to the planetographic
latitude. Using n as the £-axis, U as the *-axis, and defining y as the y-axis of the
coordinate system,
y = Uxn. (7.12-19)
The unit vector for the sub-Earth point is
jd = (jd-U, jd-yJd-n). (7.12-20)
The vector, jd = (jdxjdyjdz) is converted into the latitude and longitude of the
sub-Earth point:
t>e - sin
< -tan"1 [tan ^/(l - f)\ (7.12-22)
A^tan-^dj/jd*). (7.12-23)
The planetographic latitude and longitude of the subsolar point are found by
using the unit vector for the Sun, j, instead of the sub-Earth point unit vector, jd,
the unit vector for the Earth, in Equations 7,12-20 through 7.12-23.
The position angle and apparent distance for the subsolar point and north pole
are computed by determining the vectors specifying a coordinate system in which
the yz-plane is the plane of the sky, the jr-axis points toward the Earth, and the
>axis points toward the west as seen on the plane of the sky. In the equatorial
coordinates of date, the direction for the north pole of the Earth is j„ = (0,0, 1). So
"3QR
EXPLANATORY SUPPLEMENT
the directions for axes, (h„ Uj, uk), of the coordinate system in the ecliptic coordinate
of date are
Jt-axis, u,=jd; (7.12-24)
y-axis, Uj=jdxj„; (7.12-25)
z-axis, Ui = ajx}i. (7.12-26)
The coordinates for the north pole vector in the preceding coordinate system are
found from the equatorial coordinates by
nH = Pp(n ■ u„ n ■ uh n - u*), (7. 12-27)
where pp is the polar diameter of the planet and ft is the unit vector pointing in the
direction of the north pole found in Equation 7.12-13. The position angle of the
north pole is then
PA„ = tan- 1 (nuy I «„,). (7. j 2_28)
The length of the projection from the center of mass to the north pole on the planet
is
d = (n2la + nlz)m. (7.12-29)
Finally, the apparent distance in arcseconds is
dn = 206264. 8062[a7 (|d| - (7. 12-30)
Similarly, the position angle and apparent distance of the subsolar point is
calculated using the unit vector for the Sun, j, and the apparent radius of the
planet at the latitude of the subsolar point rather than n and the apparent polar
radius. The apparent radius of the planet at the subsolar point is derived using
Equation 7.12-9 and substituting the planetocentric latitude of the subsolar point,
The phase angle, j, of a planet is the planetocentric elongation of the Earth
from the Sun. This can be found from the dot product of j and jd:
cosi' = j-jd. (7.12-31)
The phase, k, is the ratio of the illuminated area of the disk to the total area of the
disk, as seen from the Earth:
k = 0.5(1 +COS/').
(7.12-32)
7 / PHYSICAL EPHEMERIDES OF THE SUN. MOON, PLANETS. AND SATELLITES
397
The geometric terminator is used in computing the phase and the defect of
illumination. The geometric terminator is the plane orthogonal to the direction of
the Sun passing through the center of mass of the object (see Figure 7.12.1). The
maximum geometric defect of illumination, Q, is the length of the portion of the
line segment across the face of the planet passing through the points s and e that
is not illuminated (see Figure 7.12.3). The defect calculated does not include non-
geometric effects such as the refraction of light by the planet's atmosphere. The
apparent equatorial diameter of the planet in seconds of arc is
jeq = 206264. 8062a / |d| (7. 1 2-33)
The ratio of the apparent polar diameter to the equatorial diameter that depends
on the latitude of the sub-Earth point (see Figure 7.12.2)
b'la= 1 -/(l -sin2 7), (7.12-34)
where / is the flattening of the planet and 7 is the latitude of the sub-Earth point.
The defect of illumination is t hen
!2 = 2seq[l -(1 -b't a) sin2(PAs - fi4„ + 90o)](l -*). (7.12-35)
The position angle for the defect of illumination, measured from celestial north,
is:
PAQ = PAs+\m° PAS < 1 80°,
■ PAS - 180° PAS > 180? (7.12-36)
7.2 PHYSICAL EPHEMERIS OF THE SUN
The elements used to calculate the physical ephemeris of the Sun are from Carring-
ton (1863). The values of the physical constants for the ephemeris are:
(1) mean sidereal period of rotation, Ps = 25. 38 days;
(2) inclination of solar equator to the ecliptic, / = 7?25;
(3) longitude of the ascending node of the solar equator on the ecliptic,
/? = 75?76 + 0?01397 7;;
(4) the equatorial diameter of the Sun, ds = 696000 km.
The flattening of the Sun is thought to be extremely small (/ = 0). Because the
Sun is a gaseous body, it does not rotate rigidly. The sidereal period given is for
the moan rotation rate of the Sun.
398 EXPLANATORY SUPPLEMENT
From the inclination of the Sun and the longitude of the ascending node, the
position of the north pole of the Sun is
on = 286?1300,«S0 = 63?8700. (7.2-1)
From the mean sidereal period of the Sun, the position of the prime meridian is
W = 84? 10 + 14? 1 844000 1. (7.2-2)
The heliocentric longitude and latitude of the Earth, the sub-Earth point, and
the position angle of the north pole are determined from the heliocentric position
of the Earth, re, the positions of the north pole, c*o, So, and the prime meridian
of the Sun, IV. These quantities are determined using the methods described in
Section 7.12.
The horizontal parallax is the difference in direction between the topocentric
and the geocentric coordinates of an object when the center of the object is on
the astronomical horizon. The horizontal parallax (HP), in arcseconds, is computed
using
sinffP = /?e / |re|, (7.2-3)
where Re = 6378. 137 km, the IAU equatorial radius of the Earth (Davies et ai,
1991) and re is the Earth-Sun distance.
Similarly, the semidiameter (SD) of the Sun in arcseconds is computed using
SD = 206264"8062 ds/2\re\, (7.2-4)
where ds is the diameter of the Sun. The semidiameter is then converted from
arcseconds to arcminutes and seconds.
7.3 PHYSICAL EPHEMERIS OF THE MOON
In The Astronomical Almanac data for the Moon is tabulated to an accuracy of
0?001 in longitude, latitude, and position angle.
On the average, the same hemisphere of the Moon is always turned toward the
Earth, but there are periodic oscillations or librations of the apparent position of
the lunar surface that allows about 59% of the surface to be seen from the Earth.
The librations are due partly to physical rotational librations — oscillations of the
actual rate of rotation of the Moon with respect to its mean rotation rate. A second,
much larger part of the librations are the geocentric optical librations, which are
the result of the nonuniform motion of revolution of the Moon around the Earth-
Moon barycenter. Both of these effects are taken into account in the computation
7 / PHYSICAL EPHEMEHIDES OF THE SUN, MOON, PLANETS. AND SATELLITES 399
of the Earth's selenographic longitude, Am, and latitude, (3m, and of the position
angle, C, of the axis of rotation. In The Astronomical Almanac, the contributions
from the physical librations are tabulated separately. A third contribution to the
librations, the topocentric optical librations, are a result of the difference between
the viewpoints of an observer on the surface of the Earth and the hypothetical ob-
server at the center of mass of the Earth. The topocentric optical librations may be
as large as 1° and have important effects on the apparent contour of the limb. Be-
cause the magnitude of the topocentric optical librations depend on the observer's
position, they are not tabulated in The Astronomical Almanac. Instead, a method
for estimating them is given at the end of this section and in The Astronomical Al-
manac. The Multiyear Interactive Computer Almanac (MICA), however, computes
the topocentric optical librations from rigorous formulas. The difference between
MICA and the given method for estimating the topocentric optical librations is, at
most, 1".
The selenographic coordinates of the Earth and Sun specify the point on the
lunar surface where the Earth and the Sun, respectively, are in the selenographic
zenith. The tabulated selenographic longitude and latitude of the Earth include the
total geocentric, optical, and physical librations in longitude and latitude, respec-
tively. When the libration in selenographic longitude of the Earth is positive, the
mean center of the disk is displaced eastward on the celestial sphere, exposing to
view a region of the western limb. When the libration in selenographic latitude of
the Earth is positive, the mean center of the disk is displaced toward the south,
and a region of the north limb is exposed to view.
The principal moment of inertia of the Moon closest to the Earth-Moon axis
is offset 214V2 west of the mean direction of the Earth from the Moon.
The selenographic colongitude of the Sun tabulated in The Astronomical Al-
manac is the east selenographic longitude of the morning geometric terminator. It
is determined by subtracting the selenographic longitude of the Sun from 90° and
adding 360° if the result is negative. Colongitudes of 270°, 0°, 90°, and 180° corre-
spond, approximately, to New Moon, First Quarter, Full Moon, and Last Quarter,
respectively. The age of the Moon is computed as the time elapsed, in days, since
the most recent moment at which the selenographic colongitude of the Sun was
270°.
The position angle of the midpoint of the bright limb is the same as the position
angle of the subsolar point, and it is computed in the manner described for the
subsolar point in Section 7.12. Like the other position angles, the position angle of
the midpoint of the bright limb is measured eastward around the disk from north.
The fraction of the Moon illuminated, k, is the same as the phase. This quantity
is computed as described in Section 7.1.2.
The analytic theory for the libration of the Moon used in The Astronomical
Almanac is that of Eckhardt (1981, 1982) for the physical libration of the Moon;
however, the IAU value for the inclination of the mean lunar equator to the ecliptic,
400 EXPLANATORY SUPPLEMENT
1°32'32!'7, is used instead of Eckhardt's value. In addition, there are printing errors
in Eckhardt's published values, so any attempt to use the theory must be done with
great care. Although the values of Eckhardt's constants differ slightly from those
of the IAU, the difference is of no consequence for the precision of the tabulation
except for the inclination of the lunar equator. First, the geocentric optical librations
are calculated from rigorous formulas; then, the combined geocentric optical and
physical librations are calculated using the results from the optical librations alone
Included in the calculations are perturbations for all terms greater than 0?000l
Since the apparent positions of the Sun and Moon are used in the calculations
aberration is already included, except for the inappreciable difference between the
light-time from the Sun to the Moon and from the Sun to the Earth.
The differential corrections to be applied to the tabular geocentric librations
to form the topocentric librations are derived in Atkinson (1951). The formulas for
the topocentric corrections are
A Am = - HP' sin(g - Q sec b, (7.3-1 )
AAn = + HP' cos(G - Q, (7.3-2)
AC = + sin03 + A0)l - HP1 sin Q tan 6, (7.3-3)
where Am is the selenographic longitude, /3m is the selenographic latitude, C is the
position angle of the axis of rotation, Q is the geocentric parallactic angle subtended
by the Moon, and HP' is the parallax between the point of observation on the Earth's
surface and the geocenter commonly called the topocentric horizontal parallax.
The horizontal parallax, HP, occasionally called the geocentric horizontal par-
allax, is found from the Earth-Moon distance, |rm|, and the equatorial radius of the
Earth, /?e = 6378. 137 km.
sinHP=\rm\/Re. (7.3-4)
The topocentric parallax, HP', is computed from the horizontal parallax by
using
sin HP' = sin z sin HP /( 1 - sin z sin HP), (7.3-5)
where z is the geocentric zenith angle of the Moon. The values of Q and z are
calculated from the geocentric right ascension, am, and declination, 6m, of the Moon,
using
sin z sin Q = cos $ sin h, (7.3-6)
sin z cos Q = cos<5m sin# — sin 6m cos $ cos h, (7.3-7)
cos z = sin bm sin <£ + cos 6m cos $ cos h, (7.3-8 )
j / PHYSICAL EPHEMERIDES OF THE SUN. MOON. PLANETS, AND SATELLITES 401
where <P is the geocentric latitude of the observer and h is the local hour angle of
the Moon, given by
h - LST — am, (7.3-9)
where LST is the apparent local sidereal time.
Second differences must be taken into account in the interpolation of the tabular
geocentric librations to the time of observation.
The semidiameter of the Moon in radians is simply,
SD = tan-,(«m/|rm|) (7.3-10)
where Rm = 1738 km the radius of the moon and |rm| is the Earth Moon distance.
7,4 PHYSICAL EPHEMERIDES OF THE PLANETS
In The Astronomical Almanac, the planets Mars, Jupiter, Saturn, Uranus, Neptune,
Pluto, and all of the planets in MICA have a tabulated element in their physical
ephemerides in addition to the quantities described in Section 7.1. This quantity
is the planetocentric orbital longitude of the Sun (Ls). Ls is measured eastward in
the planet's orbital plane from its vernal equinox. The planet's vernal equinox is
the point where the ascending node of the planet's equator crosses the plane of its
orbit about the Sun. The instantaneous orbital and equatorial planes are used in
computing Ls-
Ls = cos-'[(rp • rve)/ (|rp||rve|)], (7.4-1)
where rp are the ecliptic rectangular coordinates of the planet and rve are the ecliptic
rectangular coordinates of the planet's instantaneous vernal equinox. The correct
quadrant is determined from the sign of
rpxrve. (7.4-2)
Values of Ls of 0°, 90°, 180°, and 270° correspond to the beginning of spring,
summer, autumn, and winter, respectively, for the planet's northern hemisphere.
Unless otherwise noted the positions of the planetary poles and the prime merid-
ians and the planetary radii and flattenings are taken from Davies et al. (1991).
The values for the magnitude of the planets at a distance of 1 AU and at a phase
angle of 0°, as well as the value for the phase coefficient are from Harris (1961).
7.41 Mercury
The position for Mercury's pole, prime meridian, radius, V(1,0), phase coefficient ,
and flattening are given in Table 7.41.1.
402
EXPLANATORY SUPPLEMENT
Table 7.41,1
Physical Ephemeris Parameters for Mercury
= 281?01 -0?033 T
= 61?45 - OP005 T
w
= 329? 71 + 6? 1 385025 f
Am(i)
= 3.80 (// 100°) - 2,73(/7 100°)2 + 2.00(/7 100°)3
V<1,0)
= -0.36
R
= 2439.7+ 1.0 km
f
= 0
Since Mercury's orbit about the Sun is interior to the orbit of the Earth, Mer-
cury is seen to go through the entire range of phase angles. Because it passes through
such a large range in phase angle, the phase coefficient must be replaced by an an-
alytic expression (Danjon, 1954). The expression for the phase coefficient is good
for the phase angle 3° < i < 123°. For i < 3° and i > 123°, Mercury is too close to
the Sun to be observed photometrically.
The flattening is not appreciable so the equatorial and polar diameters are the
same.
The cartographic prime meridian of Mercury is defined so that the 20° meridian
passes through the center of the crater Hun Kal.
7.42 Venus
The basic parameters for Venus are given in Table 7.42.1
Venus rotates in a retrograde fashion, so the sign for the motion of the prime
meridian is negative. Like Mercury, the flattening of Venus is negligible; so the
equatorial and polar radii of Venus are the same.
Also, like Mercury, Venus' orbit is interior to the Earth's orbit, so it has a phase
angle that varies from 0° to 180°, and so an analytical function is used rather than
Table 7.42.1
Physical Ephemeris Parameters for Venus
a0 = 272?72
60 = 67? 15
W = 160?26 - 1?4813596 f
1/(1,0) = -4.29
Am(i) = 0. 09(/ / 1 00° ) + 2. 39 (/ / 1 00° )2 - 0. 65(/ / 1 00° )3
fl = 6501.9 ± 1.0km
/ = 0
7 i PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 403
a single phase coefficient (Danjon, 1949). The expression was determined for the
phase range 0?9 < i < 170P7. For r < 0?9 and i > 170?7, Venus is too close to the
Sun to be observed photometrically. Venus' change in brilliancy with rotation is
very small.
7,43 Mars
The basic parameters for Mars are given in Table 7.43.1.
Table 7.43.1
Physical Ephemeris Parameters for
Mars
ceo =
317?681 -0?108 7
6q =
52?886 - 0°061 T
W-
176? 868 + 350? 891 9830/
1/(1,0) =
-1.52
Am(f') =
0.016/
R =
3397 ±4 km
f =
0. 0065 + 0.0017
Although the phase angle of Mars can be as large as 47°, the phase relation for
its magnitude is satisfied by a single term. Mars' surface contains a large number
of visible markings. The markings cause the brilliancy of Mars to vary about 0.15
magnitudes over its period of rotation. However, the amplitude of the variation is
not fixed, and is modified by other factors such as the transparency of the atmo-
sphere from planetwide sand storms. There appears to be an annual variation as
well. All these additional factors can cause variations in the visual magnitude of up
to 0.1 magnitudes.
The cartographic prime meridian of Mars is defined so that it passes through
the center of the crater Airy-O.
7.44 Jupiter
The basic parameters for Jupiter are given in Table 7.44.1.
Multiple longitude systems are defined for Jupiter. Each system corresponds
to a different apparent rate of motion. System I applies to the mean atmospheric
equatorial rotation of the planet. System II applies to the mean atmospheric rota-
tion north of the south component of the north equatorial belt and south of the
north component of the south equatorial belt of Jupiter. System III applies to the
origin of radio emissions on Jupiter. The rotation rates for both System I and Sys-
tem II are uncertain and subject to change. The sub-Earth and subsolar longitudes
tabulated in The Astronomical Almanac are based on the System HI longitudes.
404
EXPLANATORY SUPPLEME
Table 7.44.1
Physical Ephemeris Parameters for
Jupiter
a0
= 268?05 -0?009 7
= 64P49 + 0?003 T
System I W/
- 67? 1 + 877? 900 1
System II Wlt
= 43?3 + 870?270f
System III Wm
= 284? 95 + 870? 5360000 f
tf(1,0)
= -9.25
Am(/)
= 0. 005 /
R
= 71492 ± 4km
f
= 0. 06487 ±0. 00015
Jupiter is the most luminous of the planets. Variations in the brilliancy of the
planet with rotation are small, -0.01 magnitudes, and irregular. There are also
long-period variations correlated with changes in the surface features of the planet.
7.45 Saturn
The basic parameters for Saturn are given in Table 7.45.1.
For Saturn only System III [observed rotation of radio emissions; Desch and
Kaiser (1981)] rotations are defined. Also, like Jupiter, the mean rotation rate of
the atmosphere is uncertain and subject to change. The tabulated sub-Earth and
subsolar longitudes in The Astronomical Almanac use the System III longitudes.
The previously defined System I rotation system has been found to be of little use
since the Voyager encounters with Saturn, and is no longer supported by the IAU.
Because of the brilliance of Saturn's rings, the brilliance of the planet alone is
not well known; so the combined planet-ring magnitude and surface brightness are
tabulated. The phase coefficient for Saturn includes the rings. The visual magnitude
in Table 7.45.1 is the extrapolated value for the planet alone at 1 AU and 0° phase
Table 7.45.1
Physical Ephemeris Parameters for
Saturn
<*0
= 40?58-0.°036T
«o
= 83?54-0?004 7
System III 1%
= 38?90 + 810?7939024f
1/(1-0)
= -7.19
Am(i)
= 0.044/
R
= 60268 + 4 km
f
= 0.09796 + 0.00018
7 i PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 405
ie_ There is, however, an additional term for variation in the brightness of the
planet with the aspect of the rings:
Vt = -2.60sin|ej + 1.25 sin2 e, (7.45-1)
where Vr is the visual magnitude of the rings and e is the Saturnocentric sub-Earth
latitude. There is little evidence for variations in the magnitude in the brightness
of Saturn with its rotation or over extended periods of time.
7.46 Uranus
The basic parameters for Uranus are given in Table 7.46.1,
Table 7.46.1
Physical Ephemeris Parameters for
Uranus
a0
= 257P43
h
= -15P10
System III Wm
= 203P81 - 501? 1600928 f
1/(1,0)
= -7.19
Am(/)
= 0. 0028/
R
= 25559 ± 4 km
f
= 0.02293 + 0. 00080
Like Venus, Uranus rotates in the retrograde direction. There is evidence for
variation of the brightness of Uranus with rotation, but it has not been confirmed.
A small term with a period of 84 years is expected in Uranus' magnitude due to the
flattening of the planet and the high inclination of its equator to its orbital path.
Just as for Saturn, only the rotation rate for radio emissions (System III) is defined
for Uranus.
7-47 Neptune
The basic parameters for Neptune are given in Table 7.47.1.
The phase angle of Neptune at its greatest elongation is very small, so the
variation in brightness of the planet from its phase angle is unknown. There is
some evidence of variation in the brightness of Neptune with its rotation. There is
Ho evidence of long-period variations in its brightness. Only the System III rotation
rate for radio sources is defined for Neptune.
406
EXPLANATORY SUPPLEMENT
Table 7.47.1
Physical Ephemeris Parameters for Neptune
QO = 299?36 + 0?07sinW
<50 = 43?46 - 0?51 cos N
System III Wm = 253P18 + 536°3128492 f - 0?48sin N
1/(1,0) = -6.87
R = 247641 15 km
/ = 0.0171 ±0.0013
wh@rs
/V = 357?85 + 52?316f
7.48 Pluto
The basic parameters for Pluto are given in Table 7.48.1.
The position for the north pole and the prime meridian of Pluto is derived from
Harrington and Christy (1981). The motion of the prime meridian and the position
of the pole for Pluto are given under the assumptions that the satellite's orbital
plane is coincidental with Pluto's equatorial plane and that the orbital motion of
Pluto's satellite is synchronous with Pluto's rotation. Like Uranus, the rotation of
Pluto is retrograde, and the inclination of Pluto's axis to its orbit is nearly 90°.
Because of Pluto's great distance from the Sun and its small size, its flattening is
unknown, but is presumed to be small because of its slow rotation rate.
The given brilliancy at a distance of 1 au and a phase angle of 0° is the combined
magnitude of Pluto and its satellite, Charon. The brilliancy given is from Harris
(1961); however, the brilliancy of Pluto has been changing over time as different
albedo features of the planet become visible (see Marcialis, 1988). The phase coef-
ficient given was determined by Binzel and Mulholland (1984). There is a change
in the combined magnitude with rotation of the Pluto-Charon system due to the
differing albedos of the two bodies and surface markings on Pluto. The rotational
change in magnitude, like the mean magnitude, has been changing with time as
different albedo features of the planet are exposed to the Earth in Pluto's orbit
Table 7.48.1
Physical Ephemeris Parameters for Pluto
a0 = 31 3? 02
60 = 9P09
W = 236? 77 - 56? 36231 95 f
K(1,0) = -1.01
Am = 0. 041 (
ft= 1151 ±6km
7 / PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 407
around the Sun. The rotational difference between the minimum and maximum
brilliancy was 0.11 magnitudes in 1954 (Harris, 1961) and was 0.291 magnitudes
■ 1981 (Marcialis, 1988). The period of the rotational variations is 6.3867 days.
During the eclipse season the Pluto-Charon mutual events cause changes in the
magnitude as deep as 0.50 magnitudes (Tholen et al, 1987).
7,5 PHYSICAL EPHEMERIDES OF THE SATELLITES
The disks of all the planetary satellites aside from the Moon are so small that
they cannot be directly observed, except from interplanetary probes. Therefore,
the amount of information tabulated for the physical ephemerides is much smaller
than that for the planets.
Unless otherwise stated, the values for the sidereal period of the satellites are
from Davies et al. (1991), and the value for the phase coefficients are from Harris
(1961).
7.51 Satellites of Mars
Both of Mars' two small satellites, Phobos and Deimos, are triaxial ellipsoids, with
their longest axis pointed toward the planet. The source for the reference surfaces
of these satellites is from Mariner 9 measurements by Pollack et al. (1973).
The basic parameters for the Martian satellites are given in Table 7.51.1.
Table 7.51.1
Rotation Parameters for Mars' Satellites
I Phobos: q0 = 317?68-0?108T+1?79sinM1
S0 = 52P90 - QPQ61 T - 1 °08 cos Mi
W = 35?06+1128?8445850f + 0?6644x \Q-9f - 1?42sinM, -0?78sin/W2
II Deimos: a0 = 316P65 - 0?108 T + 2P98sin M3
60 = 53P52 - 0?061 T - 1 ?78cos M3
W = 79P41 + 285? 1618970 f - 0P390 x lO"10^ - 2?58sin M3 + 0? 19C0S M3
The values for MN are
M, = 169? 51 -0? 4357640 t
M2 = 1 92? 93 + 1 1 28? 4096700 f + 0? 6644 x 1 0 " 9 12
M3 = 53?47-0?0181510f
Both of these satellites have rotation periods that are synchronous with their
orbital periods. Also, both satellites have secular accelerations in their rotation
rates which appear as the r2 terms in the motion of their prime meridians. These
acceleration terms are the result of secular changes in the mean distances of the
Martian satellites' orbits. The different signs for the quadratic terms for the motion
408 EXPLANATORY SUPPLEMENt
of the satellites' prime meridians shows that Phobos' orbit is contracting with time
while Deimos' orbit is expanding.
The masses of Phobos and Deimos are derived from a redetermination by Sjo-
gren (1983) of the perturbations of the Viking Orbiters during their close flybys of
the satellites. Information on the magnitudes of Phobos and Deimos, their albedos
and their colors are derived from Viking photometry of the satellites that has been
analyzed by Klassen, Duxbury, and Veverka (1979). Because the satellites are very
near Mars, information on the phase coefficients for both satellites and the (U — B)
color for Phobos are useless for ground-based observations and are not given.
7.52 Satellites of Jupiter
Physical ephemerides are tabulated for sixteen Jovian satellites. The most complete
and accurate information is that for the four Galilean satellites and four inner
satellites, JV (Amalthea), JXIV (Thebe), JXV (Adrastea), and JXVI (Metis).
The basic parameters for Amalthea, Thebe, Adrastea, Metis and the Galilean
satellites are given in Table 7.52.1.
These satellites all have periods of rotation that are synchronous with their
orbital periods.
The values for the visual magnitude at 1 AU and 0° phase angle, the albedos,
and the colors of the Jovian satellites are taken from Morrison (1984). The Galilean
satellites have been observed often enough that well determined phase relations have
been derived for them.
The reference surfaces for Amalthea (Smith et ai, 1979); the Galilean satellites
(Davies et ai, 1991); and JXIV, JXV, and JXVI (Thebe, Adrastea, and Metis;
Burns, 1986) were derived from measurements of the Voyager spacecraft pictures
of the satellites. Estimates of the size of the outer Jovian satellites are from Tholen
and Zellner (1984) and are based on the albedos of JVI and JVII (Himalia and
Elara).
Galilean satellite masses are calculated from the perturbation in the paths of the
Pioneer and Voyager spacecraft flybys (Campbell and Synott, 1985). The masses of
the rest of the satellites are based upon their individual sizes and the mean density
of the compositions inferred from their albedos.
Three of the Galilean Satellites have cartographic coordinates defined by surface
craters. The cartographic definitions are given in Table 7.52.2.
7.53 Satellites of Saturn
Physical ephemerides are tabulated for 17 Saturnian satellites. The rotation pe-
riods are tabulated for 13 of the Saturnian satellites. The rotational elements in
Table 7.53.1 for 11 of the satellites are supplied by Davies et al. (1991).
Table 7.52.1
Rotation Parameters for Jupiter's Satellites
XVI Metis:
ao
= 268° 05-0? 009 T
«8
= 64? 49 + 0? 003 7
W
= 302? 24+ 1221?2489660f
XV Adrastea:
a0
= 268?05 - 0?009 7
= 64? 49 + 0°003 T
w
= 5° 75 + 1206? 9950400;
V Amalthea;
= 268?05 - 0? 009 7 - 0° 84 sin J, +0?01 sin2Jt
= 64°49 + 0?003 7 - 0?36 COS J\
w
- 231 ?67 + 722? 631 4560 ? + 0?76sifl J-| - 0?01 Sin2J,
XIV Thebe:
= 268?05 - 0?009 7- 2?12sin J2 + 0°04sin 2J2
= 64? 49 + 0°003 7 - 0?91 COS J2 + 0P01 cos2J2
w
= 9? 91 + 533? 7005330 f + 1 ? 9 1 Sin J2 - 0? 04 sin 2 J%
I lo:
oca°n£ n° rtno T a. n° flQd <sin + 0°024 sin Ja
= ^OO . UD — U. Uvl* ' + U. w*r air 1 ' u. uci on i */4
k
= 64°50 + 0?003 7 + 0?040cos J3 +0°011 COS J4
w
= 200°39 + 203? 4889538 f - 0° 085 Sin J3 - 0° 022 Sin J„
Amy")
= 0.046/ - 0.0Q10/2
II Europa:
ao
= 268°08 - Q°0Q9 7+ 1?086sin J4 +0?060sin J5 + 0?015sin J6 + 0?009 sin Jj
<5o
= 64°51 +0?003 7 + 0?468cosJ4 + 0° 026 cos J5 + 0? 007 COS J6 + 0°002cos Jj
W
= 35°72+101?3747235f-0?980sinJ4 -0°054sinJs -0?014sinJ6 -0?008s
Am(()
= 0.0312/ - 0.00125/2
III Ganymede:
Q0
= 268° 20 - 0?009 7 - 0?037 sin J4 + 0?431 sin J5 + 0?091 sin J6
= 64? 57 + 0? 003 7 - 0°01 6 COS J4 + 0° 1 86 COS J5 + 0?039 COS Je
w
= 43°14 + 50°3176081 f + 0?033sinj4 -0°389sinJ5 -0°082sinJ6
Am(/)
= 0. 323/ - 0. 00066/2
IV Callisto:
Qo
n°nnQ T n0ORflcin /r- + 00ciQn <sin Jc + 0°010 sin Ja
..□n, , rtOrtnaT n°n9Qmo i_ 4. n° 9Rd ./f. — 0°004cosJq
. ^,4 y j + y. uu J 1 ■ u. Ucy coo J5 + u. u- wv-r k.^^
W
rtenOn-r . oh Q c-7-( fV7 1 C * i_rt°iifi1 ein L. 0^ Sin . — fJ°009 Sin Ja
— 259r 67 + til . oMUiior + u. uo 1 sin — u. ojo an 1 ^5 u. uiw on
Amy]
= 0. 078/ — 0, 00274/2
where the values of Jn ars
I
Jl
= 73?32 + 91473?9 7
■h
= 198?54 + 44243?8 7
•k
^ = 283D90 + 4850?7 7
•t
= 355? 80+ 1191?3 7
, = 119?90 + 262?1 7
A
; = 229?80 + 64?3 7
J)
■ = 352?25 + 2382°6 7
Ji
l = 113?35 + 6070°0
Table 7.52.2
Standard Cartographic
Longitudes for Jupiter's Satellites
Satellite Crater Meridian
Europa Cilix 182°
Ganymede Anat 128°
Callisto Saga 326°
Table 7.53.1
Rotation Parameters for Saturn's Satellites
XVIII Pan: a0 = 40?6 - O.°036 7
*0=83.°53 -0?004 7
IV = 48.° 8 +626? 04400001
XV Atlas: a0=40?58 — 0°036 T
8$ = 83?53 - 0°004 7
W = 137"8B + 598?3060OO0f
XVII Promelheus: aa = 40°5B - 0P036 7
«0=83?53 -0?004 7
W = 296° 14 + 587? 2890000 f
XVII Pandora: a0 = 40?58 - 0?036 7
60 = 83?53 -0°004 7
W= 162?92 + 572?7891000/
XI Epimetheus: aa m 40?58 - 0?0367 - 3?153sin S, + 0?086sin2S1
60 - 83?52 - 0°0047 - 0?356cosS, + 0.° 005 cos 2S,
liV - 293?87 + 518° 4907239 f + 3?133sin S, - 0?086sin 2S,
X Janus: a0 = 40?56 - 0?0367 - 1?623sin Sj + 0.°023sin2%
60 m 83? 53 - 0?004 T - 0? 1 83 cos Sv, + 0?001 cos 2Sj
W = 5B?83 + 51B?2359876( + 1°613sin% - 0?023sin2S2
I Mimas: a0 = 40°66 - 0?0367 * 13?56sin S3
60 = B3.°52 - 0?0047 - 1?53cos S3
IV = 337?46 + 381 0 9945550 f- 13°48sinS3 -44?e5sinSn
II Enceladus: u0 = 40?66 - 0?036 7
60 = 83.° 52 - 0?004 7
W = 2?82 + 262?7318996 (
III Telhys: a0 = 40?66 - 0°0367 + 9?66sin S4
*0 = 83.°52 -0?0047- 1?09cosS4
W ■ 10.° 45 * 190.°6979085 I - 9?60sin S4 + 2°23sin £9
XIII Teleslo: a0 = 50?50 - 0?036 7
60 = 84?06 - 0?004 7
W = 56° 88 + 190.° 6979330(
XIV Calypso: Qrj > 40?58 - 0°0367 + 13?943sin S5 - 1?6B6sin2S5
«0 = B3.°43 - 0?0047- 1 ? 572 COS S5 + 0? 095 cos 2S5
W = 149?36 + 190?6742373 t - 13°649sin S5 + 1?685sin2S5
IV Dione: a0 = 40?66 - 0?Q36 7
A0=B3?52 -0P004 7
W = 357.°00 + 131°5349316f
XII Helene: o0 = 40?58 - 0°0367+ 1c662sin + 0?024sin2%
60 = 83?52 - 0?0047- 0?187cosS6 + 0°C96gos2$,
W = 245?39 + 131?6174056f - 1?651 sin Sj * 0?024sin 2%
V Rhea: ct0 = 40°38 - 0?0367 + 3?10slnS,
60=83?55 -0?0047 - 0?35cosS7
W = 235°16 + 79?690047Bf- 3?08sinSV
v' "Titan: n0=36?41 - 0?0367 + 2.°66stnSa
*0=83.°94 -0?0047 - 0?30cosSs
W = 189° 64 f 22° 5769768 I - 2?64 sin Sg
VIII lapetus: a0 = 316?16 - 3?949 7
t0 = 75.°03 - 1? 143 7
W = 350° 20 + 4? 5379572 r
XI Phoebe: n0 = 355°16
60 = 68?70 - 1?143 7
W = 304?70 + 930°8338720 (
where the values of S„ are
S, = 353?32 + 75706?7 7
S5 • 28c72 + 75706?7 7
S3 = 177? 40 - 36505?5 7
St =300?00 - 7225?9 7
S5=53°59 - 8968? 8 7
Sfe = 143?3B - 10553?5 7
Se ■ 345?20 - 1016?3 7
Sa=29?80 - 52? 17
Sg =316?45 + 506°2 7
7 / PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES
411
Table 7.53.2
Standard Cartographic Longitudes
for Saturn's Satellites
Satellite
Crater
Meridian
Mimas
Palomides
162°
Enceladus
Salih
5°
Tethys
Arete
299°
Dione
Palinurus
63°
Rhea
Tore
340°
lapetus
Almeric
276°
Wisdom, Peale, and Mignard (1984) have shown that one of the major Satur-
nian satellites, Hyperion, is tumbling chaotically rather than being in synchronous
rotation. This tumbling is a result of Hyperion's triaxial shape and its rather ellip-
tical orbit with a mean distance near the 4/3 Titan resonance.
The reference surfaces for the larger satellites are derived from Smith et al.
(1982). The size of the smaller satellites are provided by their geometric albedos
(Thomas et al., 1983).
Morrison, Owen, and Sonderblom (1986) determined the magnitudes, albedos,
and colors of Saturn's satellites. lapetus is especially interesting, because the leading
hemisphere is over two magnitudes dimmer than its trailing hemisphere. Windorn
(1950) gives a parametric equation for the brightness (not the magnitude) of the
itellite of
B = 0. 57 1 - 0.429 sin a, (7.53-1)
here a is its orbital longitude, with the maximum occurring at greatest western
elongation,
Tyler et al (1981, 1982) determined the tabulated masses for Mimas, Ence-
ladus, Tethys, Dione, Rhea, Titan, and lapetus from the deflections of the Voyager
trajectories. The masses of Enceladus and Mimas are of lower accuracy and are
contradicted by resonance theories (e.g., Greenberg, 1984).
Six of the major Saturnian satellites have cartographic coordinates denned by-
surface craters. The cartographic definitions are given in Table 7.53.2.
7-54 Satellites of Uranus
The basic parameters for the 15 known satellites of Uranus are given in Table 7.54.1.
The radii for the Uranian Satellites (Smith et al, 198G) are derived from mea-
surements of Voyager photographs. The masses for the five large satellites (Viellet,
1983) are derived from mutual orbital perturbations. Smith et al. (1986) determined
Table 7.54.1
Rotation Parameters for Uranus' Satellites
XIII
XIV
Cordelia:
Ooneia:
Cressida
Juliet:
Portia:
Miranda:
Ariel:
Umbriet:
Titania:
Oberon:
q0 = 257? 31 -0?15sin(J,
i0 * -15?18*0?14cost/,
W = 127?69 - 1074?5205730 I ■
ao = 257?31 - 0?09sinU2
S0 » -15?18 + 0?09cosHj
W m 130?35 - 956?4068150 f -
0.°04sinLf,
0°02sinU7
0?08sint/a
O.°03sint^
q0 = 257?31 - 0? 16 sin Cfj
S0 = - 1 8? 1 8 + 0.° 1 6 cos U3
W = 105." 46 - 828° 391 4760 f - 0? 04 sin 1%
a0 -257?31 - 0?04sin(;4
«o - -15?18 + 0?04cos(J4
W m 59? 16 - 776? 581 6320 f - 0? 01 Sin U4
a0 = 257.°31 -0?17sinU5
*rj = — 1 5° 1 8 + 0?16cos(J5
W = 95?08 - 760P0531690 I - 0? 04 sin 1%
aa = 257?31 -0?06slnU6
60 m -15?18 + 0?06cos(/6
W = 302?56 - 730?1253660! - 0?02sin U6
«0 = 257.°31 - 0?09sfn Uj
S0 m -15?18+0?09txjsL>7
W = 25? 03 - 7O1?4865870f -
a0 = 257?31 -0?29sinL%
«o = -15?18 + 0?28cosl/8
W = 314?90 - 644?631 1260 f -
o0 =257?31 - 0?03sinU,
So = -15?ie + 0?03cost/9
W a 297? 46 - 577P3628170 f - 0?01 sin
arj = 257?31 -0?33sinU10
*o ■ -15?18 + 0?31cosU10
W = 91?24 - 472?5450690f - 0?09sin Ult>
Q0 ■ 257? 43 + 4?41 sin U„ - 0?04sin 2fJ„
Sa = -15?08 + 4?25cosiVl1 - 0?02cos2U,,
IV • 30? 70 - 254? 6906892 1 - 1? 27 sin l/12
+ 0?15sin2Ut2 + 1?15slnl/n - 0?09sln2U,,
ao = 257?43 + 0?29sin U13
60 = -15?10 + 0?28cosU13
W = 156?22 - 142?B356681 ( + Q?05 sin U12 + 0?08sin U13
a0 = 257?43 + 0?21sinl/,<
SB = -15?10 + 0?20cosU14
W - 108?05 - 86? 8688923 ( - 0°G9sin U,2 + 0?06sirl Uu
a<, = 257?43 + 0?29 sin l/,5
Sn = -1S°10 + 0?2BcosU15
W = 77?74 - 41 "3514316 ( * 0?08sin U15
a0 = 257? 43 + 0? 16 sin l/,6
60 = -15?10 + 0?16cosLf,6
W = 6?77 - 26?7394932 t + 0°04 sin U,6
where the values of Un are
= 115?75 + 54991? 87 T
u9
= 101?81
+ 12872° 63 T
[*
= 141?69 + 41887?66 T
"10
= 138?64 + 8061?81 T
(4
= 135?03 + 29927?35 7"
1*1
= 102°23
- 2024?22 T
f4
= 61?77 + 25733?59T
= 316°41 +2863° 96 T
t%
= 249° 32 + 24471? 46 T
U13
= 304°01
- 51?94T
^6
= 43°86 + 22278?41 T
Uu
= 308? 71
- 93?17 T
Uj
= 77?66 + 20289?42 T
Wis
= 340? 82
- 75? 32 T
Us
= 157?36 + 16652? 76 7
U16
= 259° 14
- 504?81 T
7 i PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 413
the magnitudes, albedos, and colors for the large Uranian satellites from Voyager 2
photometry.
7 55 Satellites of Neptune
Neptune has eight known satellites. However, the physical ephemerides of Nereid
and Triton are the only ones presently included in The Astronomical Almanac. The
basic parameters for seven of the satellites are given in Table 7.55.1.
Table 7.55.1
Rotation Parameters for Neptune's Satellites
IV
VI
VII
Naiad:
Thalassa:
Galatea:
Larissa:
VIII Proteus:
Triton:
"0
«0
w
w
■■ 299°39 + 0?70sin N - 6°49sin ftf1 +0°25sin2rt1
i 43°35 - 0?51 cosW - 4?75cosN1 +0?09cos2W1
: 254?06 + 1222? 8441209 I - 0°48sin N + 4°40sin
■■ 299? 39 + 0°70 Sin N - 0°28 Sin Afe
:43?44 - 0?51COSW - 0°21 cos W2
: 102?06+ 1155?7555612f - 0?48sin N + 0?19sin N2
0? 27 sin 2N
Despina: a0 - 299? 39 + 0? 70 sin H - 0? 09 sin W3
:43?44 -0?51COSN-0?07cOSW3
: 306°51 + 1075°7341562 ( - 0°49sin H + 0? 06 sin W3
O0 :
Q0 =
s0 =
w =
Qo
w
"0
*0
w
«o =
W :
where the values of Nn are
N = 357?85 + 52?316r
W, = 323?92 + 62606?6 7
W2 = 220°51 +55064?2 7
W3 = 354° 27 + 46564? 5 T
W4 = 75?31 +26109?4 7
W5 = 35? 36 + 14325?4 T
N6 = 142°61 +2824?6T
«7 = 177?85 + 52?316r
299°39 + 0?70sinW - 0?07sinN4
43?43 - 0?51 cosW - 0?05cosA/4
258?09 + 839? 6597686 f - 0?48 Sin W + 0?05 Sin W4
:299?38 + 0?70sinW- 0°27sin/V5
: 43°40 - 0°51 COS N — 0?20 COS N5
■■ 179°41 +649° 0534470/ - 0?48sin N + 0?19 sin Ws
: 299? 30 + 0? 70 sin N - 0? 05 sin W@
i42°90 - 0?51 cosW - 0°04oosW6
: 93? 38 + 320? 7654228 f - 0?48sin N + 0° 036 sin NB
299°36 - 32°35sm/v7-6?28stn2/v7 -2°08sin3W7 -0°74sin4W7 -
0?28sin5W7 - 0?11sin6«7 - 0?07sin7(V7 - 0?02sin8W7 -
0?01 sin 9^7
41?17 + 22?55 cos Nj + 2? 1 0 cos 2^ + 0° 55 cos 3/V? +0° 16 cos 4^ +
0? 05 cos 5 Nj + 0°02cos6W7. + 0°01 cos7«7
296° 53 - 61?2572637/ + 22D25sinrV7 + 6.°73sin2W7 +2?05sin3/V7 +
0?74sin4rV7 + 0?28sin5W7. + 0?11sin6W7 + 0?05sin7W7 +
0°02sin8rV7+0°01sin9rV7
414 EXPLANATORY SUPPLEMENT
The mass, radius, and colors of Triton are determined from Voyager 2 data
Little is known of Nereid except its magnitude at 1 AU and 0° phase angle (Mor-
rison and Cruikshank, 1974). The diameter is based on an assumed albedo of 0.4
Recent studies by Williams et al. (1991) of Nereid indicate that it may have
non-synchronous rotation.
7.56 The Satellite of Pluto
Charon, Pluto's only known satellite, is assumed to orbit in the plane of Pluto's
equator and in synchronous rotation. The position of its north pole and prime
meridian are given in Table 7.56.1.
Table 7.56.1
Rotation Parameters for Pluto's Satellite
a0 =312?02
<50 = 9?09
W = 56° 77- 56? 3623195?
The mass, diameter, and magnitude of Charon are based on observations of
Charon-Pluto eclipses (Dunbar and Tedesco, 1986).
7.6 PHYSICAL EPHEMERIDES OF THE ASTEROIDS
At present, the only physical data published in The Astronomical Almanac for the
asteroids are H, the absolute visual magnitude at a distance of 1 AU and 0° phase
angle; G, the slope parameter that depends on the albedo; the visual magnitude at
opposition; and the estimated diameters for 139 asteroids. These data are supplied
by the Institute for Theoretical Astronomy, St. Petersburg, Russia. Additional in-
formation on the asteroids can be found in the Institute for Theoretical Astronomy
publication, Minor Planets.
The visual magnitudes of the asteroids are based on the work of Bowell, Harris,
and Lumme (1988).
Before 1970, the diameter measurements were based on visual micrometer mea-
surements or occupation observations. Reliable measurements of asteroid radii {e-g-<
Dollfus, 1971; Millis et al, 1987) existed for only five asteroids: Ceres, Pallas, Juno,
Vesta, and Eros. Since then, improved estimates of the diameters have resulted fr°*
development of the radiometric and polarimetric methods. The diameters publishe
in The Astronomical Almanac use the radiometric observations of Morrison (1977)-
The radiometric method is simple in its physical principle, but compleX t0
discuss quantitatively. The visible brightness of an asteroid is proportional to the
? i PHYSICAL EPHEMERIDES OF THE SUN. MOON, PLANETS, AND SATELLITES 41 5
product of its geometric albedo and the square of its mean diameter. If one of these
factors is known, or a relation between the two factors can be determined, the other
factor can be determined. At thermal infrared wavelengths, the dependence of the
brightness on albedo is complementary to the dependence at visible wavelengths.
For example, take two objects at the same distance with the same visible magni-
tude. If one is dark and large, and the other is small but highly reflecting, then in
infrared emitted radiation, the first object will be much brighter than the second
one, because of both its larger size and its higher temperature. The complemen-
tary nature of brightness with albedo in the two spectral regions is the basis for
the radiometric method of determining diameters. The equations relating the ob-
servable parameters (visible and infrared magnitudes) to the unknowns (diameter
and geometric albedo) require complicated assumptions to be made concerning the
temperature distribution over the surface of the asteroids and cannot be solved
analytically.
The V magnitude for an object at phase angle i and R = I AU from the Sun is
related to the radius, r, and geometric albedo, pv by
V = ys - 5 log r - 2. 5 \ogpv -2,5 log f3v(i) + log R, (7.6-1)
where Vs is the magnitude of the Sun at 1 AU and (3v(i) is the phase coefficient in
intensity units, normalized to unity at i — 0°. The brightness of the same object at
10 fim (N magnitude) is
N = KN - 5 log r - 2. 5 \og(IN) - 2. 5 log eN(i), (7.6-2)
where Kn is a constant determined by the absolute calibration of the JV-magnitude
system; lN is the infrared surface brightness in the N band, obtained by averaging
the emission from each surface element; and eN{i) is the effective emissivity in the
N band at phase angle i. Note that eN{0) can be greater than unity to allow for
Peaking of the infrared emission at small phase angles, but at the same time eN(i)
must be substantially less than unity at large i to satisfy the total heat balance,
ft h is assumed that the planet is seen at zero phase and the surface temperature
ls every where in equilibrium with the absorbed solar energy, the distribution of
temperature over the surface will be given by
T(a) = T0R-m[(l -B)/e]l/4cos"V (7.6-3)
ghere To is the equilibrium blackbody temperature at 1 AU (= 295°K), B is the
°nd albedo, e is the average emissivity, and a is the local solar zenith angle.
41 6 EXPLANATORY SUPPLEMENT
A first-order solution may be obtained if the infrared brightness is assumed to
be proportional to the absorbed energy:
IN = TlR~2(l -B), (7.6-4)
where T\ is the mean surface temperature for the body. When this substitution is
made and Equation 7.6-2 is subtracted from Equation 7.6-1, the result is
V - N = { Vs - KN) - 2. 5 logWv I eNT\ ) - 2. 5 log[pv / (1 - qpv)], (7.6-5)
where q is the phase integral adjusted to correct for the difference between pv and its
average or bolometric equivalent. To solve for pv from the observables V and N, it is
necessary to know the photometric parameters Vs, KN, and T\ (or its more general
equivalent 7q). The ratio of the phase functions in both the visible and infrared
must be known also. These four quantities can be obtained from observations over
a range of phase angles. Finally, the phase integral, q, must be specified to obtain
a solution. The values of all these parameters must either be assumed or treated as
variables in a calibration of the method based upon objects of known diameter.
Since the phase integral is not generally known, it is important to consider how
sensitive the derived values of r, pv, and B are to changes in q. Of particular signif-
icance is the behavior of the solutions for the low-albedo objects that predominate
in the asteroid belt. In Equation 7.6-5 as pv approaches zero, the final term ap-
proaches —2. 5 log/jy, so that for dark surfaces, the solution is virtually independent
of the value of q. However, a similar exercise for B shows that, even at very low
albedos, the final value of the Bond albedo remains inversely proportional to the
assumed phase integral.
In this simplified solution, the expression for the radius is
5 \og(r/ R) = KN -N -2.5 log{£NTi) -2.5 log(\ - B). (7.6-6)
This expression does not contain any of the parameters of the photometric system
including the observed V magnitude. However, the derived radius is sensitive to
the infrared parameters, Ajv, £n, and T\, in addition to the observed N magnitude.
Because the albedo enters only via the (1 - B) term, r is only weakly dependent on
the albedo, and for low albedos the final term is negligible. For this reason, a much
better estimate of the diameter can be made from a measurement of N alone than
from one of V alone, even when the infrared magnitude is less well-determined than
the visual magnitude. Furthermore, even a very rough value of V, when combined
with a measurement of N, is sufficient to define a reasonably precise diameter.
One of the assumptions of the radiometric models is that the local surface tem-
perature is approximately in instantaneous equilibrium with the incident sunlight-
Infrared observations of Phobos (Gatley et ai, 1974), Eros (Morrison, 1976), and
7 / PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 417
the Galilean satellites (Morrison and Cruikshank, 1973; Hansen, 1973) have shown
that all these objects have surface thermal conductivities lower than that of the
Moon. As a result of these low conductivities, the surfaces all respond very rapidly
to changes in sunlight; so they remain close to equilibrium even for a rapidly rotat-
ing asteroid such as Eros, with a period of less than 6 hours. So the basic assumption
of thermal equilibrium appears to be valid.
7.7 REFERENCES
Atkinson, R.d'E. (1951). "The Computation of Topocetitric Librations" Mon. Not. Roy.
Astr. Soc. Ill, 448-454.
Binzel, R.P. and Mulholland, J.D. (1984). "Photometry of Pluto During the 1983 Oppo-
sition: A New Determination of the Phase Coefficient" Astron. J, 89, 1759-1761.
Burns, J. (1986). "Some Background About Satellites" in Satellites J. A. Burns and M.S.
Matthews, eds. (University of Arizona Press, Tucson, AZ), p. 1-38.
Campbell, J.K. and Synott, S.P. (1985). "Gravity Field of the Jovian System from Pioneer
and Voyager Tracking Data" Astron. J. 90, 364-372.
Carrington, R.C. (1863). Observations of Spots on the Sun (Williams and Norgate, Lon-
don).
Danjon, A. (1949). "Photometrie et colorimetrie des planetes Mercure et Venus" Bull.
Astron. 14, 2e Serie, 315-345.
Danjon, A. (1954). "Magnitude et albedo visuels de la plane te Mercure" Bull, Astron.
17, 2e Serie, 363.
Davies, M.E., Abalakin, V.K., Brahic, A., Bursa, M., Chavitz, B.H., Lieske, J.H., Seidel-
mann, P.K., Sinclair, A.T., and Tjuflin, Y.S, (1991). "Report of the IAU/IAG/COSPAR
Working Group on Cartographic Coordinates and Rotational Elements of the Planets
and Satellites: 1991."
Davies, M.E., Hauge, T.A., Katayama, F.Y., and Roth, J. A. (1979). Control Networks for
the Galilean Satellites: November 1919 The Rand Corporation, R-2532-JPL/NASA.
Desch, M.D. and Kaiser, M.L. (1981). "Voyager Measurement of the Rotation Period of
Saturn's Magnetic Field" Geophys. Res. Lett. 8, 253-256.
Dollfus, A. (1971). "Diameter Measurements of Asteroids" in Physical Studies of Minor
Planets T. Gerhels, ed. NASA SP-267 (NASA), pp. 25-32,
Dunbar, R.S. and Tedesco, E.F. (1986). "Modeling Pluto-Charon Mutual Eclipse Events:
L First-Order Models" Astron. J. 92, 1201-1209.
Duncombe, R.L., Klepczynski, W.J., and Seidelmann, P.K. (1974). "The Masses of the
Planets, Satellites, and Asteroids" Fund. Cos. Phy. 1, 119-165.
Eckhardt, D.H. (1981). "Theory of the Libration of the Moon" The Moon and the Planets
25, 3-49.
Eckhardt, D.H. (1982). "Planetary and Earth Figure Perturbations in the Libration of the
Moon" in High-Precision Earth Rotation and Earth-Moon Dynamics O. Calame, ed.
(Reidel, Dordrecht, Holland), pp. 193-198.
Ga*ley, I., Kieffer, H., Miner, E., and Neugebauer, G. (1974). "Infrared Observations of
Phobos from Mariner 9" Astrophys. J. 190, 497-503.
Greenberg, R. (1984). "Orbital Resonances among Saturn's Satellites" in Saturn T. Ger-
hels and M.S. Matthews, eds. (University of Arizona Press, Tucson, AZ), pp. 593-608.
418
EXPLANATORY SUPPLEMENT
Hansen. O.L. (1973). "Ten Eclipse Observations of Io, Europa, and Ganymede" Icarus lg
237 246.
Harris, D.L. (1961). "Photometry and Colorimetry of Planets and Satellites" in Planets
and Satellites G.P. Kuiper and B.A. Middlehurst, eds. (University of Chicago Press
Chicago), pp. 272-342.
Kuiper, G.P. (1954). "Report of the Commission for Physical Observations of the Planets
and Satellites" Trans. IA U 9, 250.
Lumme, K. and Bowell, E. (1981). "Radiative Transfer of Atmosphereless Bodies, Ip
Interpretation of Phase Curves" Astron. J. 86, 1705-1721.
Lumme, K., Karttunen, H., Bowell, E., and Poutanen, M. (1986). "Inversion of Aster-
oid Light Curves Using Spherical Harmonics" in Asteroids, Comets, Meteors 7/C.-I.
Langerkvist, B.A. Lindblad, H. Lundstedt, and H. Rickman, eds. (Uppsala University,
Uppsala, Sweden) pp. 55-59.
Lumme, K., Karttunen, H., Piironen, J., and Bowell, E. (1986). "Simultaneous Solutions
for Pole, Shape, and Albedo Variegation of an Asteroid" Bull. Am. Astron. Soc. 18,
801.
Marcialis, R.L. (1988). "A Two Spot Albedo Model for the Surface of Pluto" Astron. J.
95, 941 947.
Millis, R.L., Wasserman, L.H., Franz, O.G., Nye, R.A., Oliver, R.C., Kreidl, T.J., Jones,
S.E., Hubbard, W„ Lebofsky, L., Goff, R., Marcialis, R., Sykes, M., Frecker, J.,
Hunten, D., Zellner, B., Reitsema, H., Schneider, G., Dunham, E., Klavetter, J.,
Meech, K., Oswalt, T., Rafeit, J., Strother, E., Smith, J., Povenmire, H-, Jones, B.,
Kornbluh, B., Reed, L., Izor, K., A'Hearn, M.F., Schnurr, R., Osborn, W., Parker,
D. , Douglas, W.T., Beish, J.D., Klemola, A.R., Rios, M., Sanchez, A., Piironen, J.,
Mooney, M., Ireland, R.S., and Leibow, D. (1987). "The Size, Shape, Density, and
Albedo of Ceres from Its Occultation of BD + 8°417" Icarus 72, 507-518.
Morrison, D. (1976). "The Diameter and Thermal Inertia of 433 Eros" Icarus 28, 125-132.
Morrison, D. (1977). "Asteroid Sizes and Albedos" Icarus 31, 185-220.
Morrison, D. (1982). "Introduction to the Satellites of Jupiter" in Satellites of Jupiter D.
Morrison, ed. (University of Arizona Press, Tucson, AZ), pp. 3-43.
Morrison, D. and Cruikshank, D.P. ( 1973). "Thermal Properties of the Galilean Satellites"
Icarus 18, 224-236.
Morrison, D, and Cruikshank, D.P. (1974). "Physical Properties of the Natural Satellites"
Spac. Sci. Rev. 15, 641-739.
Morrison, D., Owen, T., and Sonderblom, L.A. (1986). "The Satellites of Saturn" in Satel-
lites J. A. Burns and M.S. Matthews, eds. (University of Arizona Press, Tucson, AZ),
pp. 764-801.
Pollack, J.B., Veverka, J., Noland, M., Sagan, C, Duxbury, T.C., Acton, C.H., Jr., Born,
G.H., Hartmann, W.K., and Smith, B.A. (1973). "Mariner 9 Television Observations
of Phobos and Deimos, 2" J. Geophys. Res. 78, 4313-4326.
Sjogren, W.L. (1983). "Planetary Geodesy" Rev. Geophys. Space Phys. 21, 528-537.
Smith, B.A., Sonderblom, L., Batson, R., Bridges, P., Inge, J., Masursky, H., Shoemaker,
E. , Beebe, R., Boyce, J., Briggs, G., Bunker, A., Collins, S.A., Hansen, C.V., Johnson,
T.V., Mitchell, J*L., Terrile, R.J., Cook, A.F., II, Cuzzi, J., Pollack, J.B., Danielson,
G.E., Morrison, D., Owen, T., Sagan, C, Ververka, J., Strom, R., and Suomi, V.E-
(1982). "A New Look at the Saturn System: The Voyager 2 Images" Science 215.
504-537.
7 / PHYSICAL EPHEMERIDES OF THE SUN, MOON, PLANETS, AND SATELLITES 41 9
gjuith, B.A., Sonderblom, L.A., Beebe, R., Bliss, D., Boyce, J.M., Brahic, A., Briggs,
G.A., Brown, R.H., Collins, S.A., Cook, A.F., II, Croft, S.K., Cuzzi, J.N., Daniel-
son, G.E., Davies, M.E., Dowling, T.E., Godfry, D., Hansen, C.J., Harris, C, Hunt,
G.E., Ingersoll, A. P., Johnson, T.V., Krauss, R.J., Masursky, H., Morrison, D., Owen,
T., Plescia, J., Pollack, J.B., Porco, CP., Rages, K., Sagan, C, Shoemaker, E.M.,
Sromovsky, L.A., Stoker, C, Strom, R.G., Suomi, V.E., Synott, S.P., Terrile, R.J.,
Thomas, P., Thompson, W.R., and Veverka, J. (1986). "Voyager 2 in the Uranian
System: Imaging Science Results" Science 233, 43-64.
Smith, B.A., Sonderblom. L.A., Beebe, R., Boyce, J., Briggs, G., Carr, M., Collins, S.A.,
Cook, A.F., II, Danielson, G.E., Davies, M.E., Hunt, G.E., Ingersoll, E., Johnson,
T.V., Masursky, H., McCauley, J., Morrison, D., Owen, T., Sagan, C, Shoemaker,
E.M., Strom, R., Suomi, V.E., and Veverka, J. (1979). "The Galilean Satellites and
Jupiter: Voyager 2 Imaging Science Results" Science 206, 927-950.
Thomas, P., Veverka, J., Morrison, D., Davies, M., and Johnson, T.V. (1983). "Saturn's
Small Satellites: Voyager Imaging Results" J. Geophys. Res. 88, 8743-8754.
Tholen, D.J. and Zellner, B. (1983). "Multi-color Photometry of the Outer Jovian Satel-
lites" Icarus 53, 341-347.
Tholen, D.J., Buie, M.W., and Swift, C.E. (1987). "Circumstances for Pluto-Charon Mu-
tual Events in 1988" Astron. J. 94, 1681-1685.
Tyler, G.L., Eshleman, V.R., Anderson, J.D., Levy, G.S., Lindal, G.F., Wood, G.E., and
Croft, T.A. (1981). "Radio Science Investigations of the Saturnian System: Voyager
1 Preliminary Results" Science 212, 201-206.
Tyler, G.L., Eshleman, V.R., Anderson, J.D., Levy, G.S., Lindal, G.F., Wood, G.E., and
Croft, T.A. (1982). "Radio Science with Voyager 2 at Saturn: Atmosphere and Iono-
sphere and Masses of Mimas, Tethys, and Iapetus" Science 215, 553-558.
Viellet, C. (1983). "De TObservation et du Mouvement des Satellites d'Uranus" Ph.D.
Thesis. (Universite de Paris, Paris).
Williams, LP., Jones, D.H.P, and Taylor, D.P. (1991). "The Rotation of Nereid" Mon.
Not. R. Astr. Soc. lp-2p.
Wisdom, J., Peale, S.J., and Mignard, F. (1984). "The Chaotic Rotation of Hyperion"
Icarus 58, 137 152.
CHAPTER
Eclipses of the Sun and Moon
by Alan D. Fiala and John A. Bangert
8.1 INTRODUCTION
All bodies orbiting in the solar system cast a shadow away from the Sun. In the most
general terms, an eclipse occurs when the shadow of one body falls upon another
and temporarily blocks out a portion of the solar illumination. An eclipse is distinct
from an occultation, which is a block in the line of sight. Before this distinction was
understood, however, a long history of observation from Earth created different
names for the phenomena, depending on what objects were involved. The shadow
of the Moon falling upon the Earth was called a solar eclipse; the shadow of the
Earth falling upon the Moon was called a lunar eclipse; and the shadow of a planet
falling upon the Earth was called a transit (across the Sun). By strict application
of these definitions, a solar eclipse is really an occultation.
The basic methods by which eclipses of the Sun and Moon and transits of Venus
and Mercury are predicted and described were introduced by Bessel and extensively
developed by Chauvenet (1891). The discussion here will be limited to describing
the calculation of data that are necessary to observe any of these phenomena from
the Earth.
Bessel's method of calculating eclipse predictions is based on the following con-
cept: A body casts a shadow, the axis of which is a line extended from the center of
the Sun through the center of the body. Observations, by necessity, are constrained
to the topocentric coordinate system. However, the geometry and kinematics are
easier to describe in a system oriented to the cone of the shadow — that is, one in
which the z-axis is either parallel to, or coincident with, the shadow axis, and the
■^y-plane passes through the center of the body on which the shadow falls. This plane
is also known as the Besselian plane. A description of the progress of the eclipse is
calculated in this coordinate system, and then is transformed to a topocentric sys-
tem as necessary. Elements describe the geometry of the configuration as a function
°f time; circumstances describe distinct, observable events by time and place.
421
422 EXPLANATORY SUPPLEMENT
Diagrams in treatises on eclipses by necessity are exaggerated to illustrate
points. Hence, the diagrams tend to disguise the fact that, in reality, the bodies
involved are very small, far apart, and moving relatively slowly. Thus, the shadows
cast are long and slender, there are many small angles, and the relative short-term
motions are very nearly linear. These qualities were used to advantage by people
equipped only with tables and a calculator, as it was necessary and adequate to use
approximations and intermediate trigonometric constructs. In the era of high-speed
programmable computers, the use of approximations and simplifying constructs is
reduced but not eliminated. Also, it is helpful to newer generations of astronomers
to discuss spherical trigonometry by incorporating more vector and matrix notation
instead of direction cosines.
8.1 1 Eclipse Data Available from the Nautical Almanac Office
8.111 International Agreements Ever since astronomical ephemerides and alma-
nacs have been published, eclipse information has been included. This was a natural
action because the extensive calculations required are based on predicted ephem-
erides. In modern times, the situation is much eased, but the Nautical Almanac
Office continues to provide information on every eclipse of each year. The infor-
mation conforms to IAU standards and is not changed without adequate notice
to the International Astronomical Union. This may cause some misunderstandings
(e.g., in the value of /t, described in Section 8.12), or occasional difficulties when a
convention changes (for instance, longitude is now considered positive to the east
of Greenwich). Some changes that improve production without changing the infor-
mation are adopted by decisions within the Nautical Almanac Office, such as the
change in the solar eclipse map curves to an older convention, which made mechan-
ical computer-controlled plotting feasible. Other special publications for individual
eclipses are designed to assist the scientific observer as much as possible. Since
1949, at the request of the appropriate Commissions of the International Astro-
nomical Union, the principal circumstances of all solar eclipses have been routinely
calculated and made available several years in advance of the publication of the an-
nual almanac. This working agreement continues and is carried out by the special
publications described in Section 8.113.
8.112 Data Provided in The Astronomical Almanac Elements and general circum-
stances are given in The Astronomical Almanac for all solar and lunar eclipses,
including the penumbra! lunar eclipses that occur during the year. For the solar
eclipses, maps are given from which approximate local circumstances may be ob-
tained for any particular place, and the Besselian elements are tabulated at intervals
often minutes for the calculation of accurate local circumstances for any point on or
above the surface of the Earth. For total or annular solar eclipses, the latitudes and
longitudes of points on the central line and on the northern and southern limits, the
8 / ECLIPSES OF THE SUN AND MOON 423
duration of the total or annular phase, and the altitude of the Sun on the central
line are tabulated at intervals of ten minutes or less throughout the eclipse. For
lunar eclipses, the circumstances and their times are the same for all parts of the
Earth; any particular phase is visible from any point at which the Moon is above
the horizon.
For years in which a transit of Mercury or Venus occurs, general circumstances
and a map are given, by which approximate local circumstances may easily be
estimated. Formulas are also provided for the computation of more accurate local
circumstances.
The information that appears in The Astronomical Almanac is prepared a min-
imum of three years before the event. A provisional value of AT to the nearest whole
second is assigned, so that the time argument is provisional Universal Time (UT)
or, more precisely, provisional UTl (see Chapter 10). More details are given in
Sections 8.12 and 8.363.
The predictions given in The Astronomical Almanac take no account of the
effects of refraction, although refraction may be significant for the reduction of
observations intended to give precise positions. The Besselian elements, however,
are rigorously independent of refraction.
8.113 Data Provided in Other Publications Material from the section of The As-
tronomical Almanac on solar and lunar eclipses and transits is also published, in
advance of publication of The Astronomical Almanac itself, in the small booklet
Astronomical Phenomena, issued annually by the Nautical Almanac Office, U.S.
Naval Observat ory.
In 1949, the U.S. Naval Observatory began a publication series called the U.S.
Naval Observatory Circulars. The series transmits astronomical data resulting from
continuing programs but not available in any other publications. Two of the con-
tinuing programs that have contributed many issues to this series are solar eclipse
predictions and sunspot counts. There are two kinds of solar eclipse circulars. First,
there are circulars issued well in advance that give just the principal circumstances
of all solar eclipses in a multiple-year period. These are listed in Table 8.113.1.
Second, the Observatory issues circulars for specific solar eclipses; these circulars
contain extensive information for observers to choose a site and calculate accu-
rate local circumstances. Before the circulars were begun in 1949, this information
was issued in Supplements to the annual American Ephemeris. Until 1991, it was
the policy of the U.S. Naval Observatory to issue a circular for each total solar
eclipse, with a very few exceptions, and for selected annular solar eclipses that have
some observationally interesting features. Typically these circulars contained gen-
eral information and circumstances, Besselian elements, a discussion of anticipated
meteorological conditions along the eclipse path, local circumstances for selected
^cations, the lunar limb profile and limb corrections, path of the central phase tab-
ulated by time and longitude at sea level and at flying altitudes, and detail maps of
424
EXPLANATORY SUPPLEMENT
Table 8.113.1
U.S.N.O. Solar Eclipse Circulars
Lrir. no.
Ha to
L/dlc
Ti+la
1 1 lie
1
iy4y July i a
—
Tracks of total solar eclipses in 1952, 1953, 1954
o
£
jy^-y oepiemuer iy
Tracks of total solar eclipses in 1955, 1956, 1957
16
lyou May 4
Tracks of total solar eclipses in 1958, 1959, 1960
40
1952 December 8
Annular and partial solar eclipses in 1955, 1956, and 1957,
Non-central total eclipses of 1 957
i yo4 August d\j
Annular eclipses in 1958, 1959
59
1 955 June 28
Solar eclipses, 1960-1963
85
1958 April 16
Solar eclipses, 1 963-1 967
89
1960 June 15
Solar eclipses, 1 968-1 970
101
1964 July 29
Solar eclipses, 1971-1975
113
1966 December 15
Solar eclipses, 1976-1980
142
1973 April 20
Solar eclipses, 1981-1990
170
1986 November 7
Solar eclipses, 1991-2000
Note: No. 170 was the first to contain eclipse maps. The others listed only numerical data.
the portion of the path over land. The U.S. Naval Observatory planned to release
each circular 12 to 18 months in advance of a particular eclipse and to incorporate
the best value of AT available at the time of preparation. However, the lead time
and contents were subject to regular review with respect to meeting the reported
needs of the scientific community. In 1992, the eclipse circulars were discontinued.
8.12 Corrections to the Ephemerides
The basic quantities for calculations of solar and lunar eclipses are: the apparent
right ascension (as), declination (c~s), and distance {Rs) of the Sun, and the apparent
right ascension (am), declination (6m), and distance (Rm) of the Moon, for every hour
of TDT during the eclipse; and the ephemeris sidereal time at 0h TDT for the day of
the eclipse and for the following day. The apparent places of the Sun and the Moon
are computed rigorously using the same ephemerides for the Sun and the Moon as
are used for The Astronomical Almanac, and using the reduction methods described
elsewhere in this book (see Chapter 3). At the time of this writing, the ephemerides
are based upon the DE200/LE200 produced by the Jet Propulsion Laboratory. The
ephemeris sidereal time at 0h TDT, which is the local apparent sidereal time on the
ephemeris meridian, is the same numerically as the (Greenwich) apparent sidereal
time at 0h UT, as tabulated in The Astronomical Almanac.
Gravitational ephemerides refer to the positions of the centers of mass of the
bodies concerned. Eclipses, however, are governed by the positions of the centers of
figures of the Sun and the Moon. The center of figure of the Moon does not coincide
8 / ECLIPSES OF THE SUN AND MOON 425
•with its center of mass; to allow for this difference, empirical corrections are made to
the mean equatorial coordinates of the Moon, referred to the center of mass of the
Earth. Unpublished investigations have shown that the corrections depend upon
the lunar librations at the time of an eclipse. The range of the corrections is less
than 0"5, and for the calculations in The Astronomical Almanac, only the minimum
corrections are applied. These are AA = +0"50 and A/3 - — 0"25 for the LE200 lunar
ephemeris. The corrections are made during the apparent place computation by
rotating the mean equatorial geocentric coordinates of the Moon into the ecliptic
system, applying the corrections in the ecliptic system, and rotating back into the
equatorial system. The corrections are applied as constants for each eclipse.
The semidiameters of the Sun and the Moon used in the calculation of eclipses
do not include irradiation. The adopted semidiameter of the Sun (ss) is computed
directly from the ephemeris of the Sun, thereby using the ephemeris constants.
However, the apparent semidiameter of the Moon is calculated by putting its sine
equal to k sin 7rm, where 7rm is the horizontal parallax and k, the ratio of the Moon's
radius to the equatorial radius of the Earth, is an adopted constant. The IAU
adopted a new value of k (k = 0. 2725076) in August 1982. Before that time, two
different values of k were used in computing a solar eclipse. The value taken for
the lunar ephemeris was used for the most part, but in practice it was too large for
total phase in total solar eclipses because of the effects of the rough lunar limb. As
an approximate correction for this effect, a smaller value of k was adopted solely
for calculating duration on the central line of total solar eclipses. This smaller
value caused numerical inconsistencies as well as misunderstandings on the part
of users of the information. Therefore, when the new value was adopted in 1982
to conform to the value used in occultation predictions, it was agreed implicitly
that limb effects are no longer accounted for, but are averaged, and if an observer
considers them to be important, then corrections must be calculated and applied
separately. It is possible for these effects to advance or retard predicted second-
er third-contact times on the central line by as much as two seconds apiece. The
same limb corrections can be applied in the case of annular solar eclipses, or even
lunar eclipses, but the effects will not be observable. It is only in the case of a total
solar eclipse that the presence or absence of any light from the solar disk is easily
detected and is hence critical to delicate observations.
The calculation of the occurrence of eclipses does not depend upon the time
scale for the portions restricted to the spatial relationships of the Earth, Moon, and
Sun. However, for calculation of phenomena as they are related to the surface of the
Earth for time and place of visibility, it is necessary to use Universal Time (UT1) in
the best approximation possible. This approximation is achieved by adjusting the
ephemerides at the start of the calculations. The predictions in the Astronomical
Almanac were given in ephemeris time (ET) until 1981, when AT had become large
enough that it was deemed more useful to publish the predictions in provisional
Universal Time (see Section 8.363 for details).
426 EXPLANATORY SUPPLEMENT
8.2 THE OCCURRENCE OF LUNAR AND SOLAR ECLIPSES
8.21 Overview
The purpose of this section is to develop conditions that can be used to determine
whether an eclipse will occur at a particular conjunction or opposition of the Sun
and Moon. The approach is, first, to develop a general expression for the geocentric
least angular separation of the Moon and either the Sun or the antisolar point, when
one of the latter is in the vicinity of one of the nodes of the Moon's orbit. Second,
expressions for the apparent sizes of the shadows are found. FinaUy, discrimination
tests are established. Specifically, to develop the conditions for a lunar eclipse the
apparent sizes of the umbral and penumbral shadows of the Earth are compared
to the geocentric least angular separation of the Moon and the antisolar point.
To develop the conditions for a solar eclipse, expressions for the geocentric least
angular separation of the Sun and the Moon are derived from the geometry of
each particular type of solar eclipse and are compared with the geocentric apparent
angular semidiameters.
8.22 Geocentric Least Angular Separation
Consider the Sun and the Moon at a time of conjunction or opposition in longitude.
In order for an eclipse to occur, the Sun (or the antisolar point) and the Moon
must each be near the same one of the nodes of the Moon's orbit. Thus, the ecliptic
latitude of the Moon (/?m) will be small, and plane geometry may be used.
Ecliptic
Moon's orbit
Figure 8.22.1
Geometric construct for determining whether any kind of eclipse will occur when the Moon is near
one of the nodes of its orbit on the ecliptic. In this figure, N is the descending node as seen from
outside the orbit, looking toward Earth. Let S represent the Sun or its antisolar point, and M the Moon,
at conjunction or opposition, respectively, when the apparent separation is the lunar latitude 0m- At
some later time when the two points have moved to S', M' , then the instantaneous separation is a.
a I ECLIPSES OF THE SUN AND MOON 427
This situation is illustrated in Figure 8.22. 1. This is a general representation; (3m
may t>e positive (above the ecliptic) or negative (below the ecliptic). Correspond-
'nfflv /V may De either the ascending or descending node of the lunar orbit. For a
lunar eclipse, S represents the antisolar point at opposition. For a solar eclipse, S
represents the center of the Sun at conjunction. For either case, M represents the
center of the Moon. Take the absolute value of Pm, which is the latitude of the
Moon at opposition for a lunar eclipse, or at conjunction for a solar eclipse. The
primed values, M' and 5', represent positions at a time when angle SMS' = 7, and
the angular separation of M' and 5" is ex. I is the inclination of the Moon's orbit to
the ecliptic.
We wish now to derive an expression for a in terms of known quantities. Define
a quantity A, a ratio of the longitudinal motions:
SP Moon's motion in longitude
A = — — : : : — - (5.22—1)
SS' Sun's motion in longitude
From the geometry:
where
<r2 = (S'P)1 + {M'Pf, (8.22-2)
S'P = SP - SS' = ( A - 1 )/3m tan 7. (8.22-3)
M'P = NP tan / = (SN - SP) tan / = /3m(l - A tan 7 tan/). (8.22-4)
Thus,
a2 = /32[(A - l)2tan27 + (l - Atan/tan7)2]. (8.22^5)
In order to find the value of 7 at which er2 is a minimum, take the derivative of
Equation 8.22-5 with respect to 7 and set it equal to zero. After some algebra, the
result is:
tan 7
Atan/ (8.22-6)
(A - 1)2 + A2tan2/"
Substitute Equation 8.22-6 into Equation 8.22-5 to obtain:
a = /3m(A - 1)[(A - l)2 + A2 tan2 (8-22-7)
This equation may be simplified by the following change of variables. Let
tan /' = -r—. - tan /. (8.22-8)
A — 1
428
EXPLANATORY SUPPLEMENT
1' is the angle at which the moving Moon would appear to intersect the eclmH
from the point of view of an observer on the moving point S. Then
<T=/?mCOs/'- (8.22-9)
Equations 8.22-7 and 8.22-9 give expressions for the geocentric least angular seDa
ration between the Moon and the Sun at conjunction, or the geocentric least an JLr
separation of the Moon and the antisolar point at opposition of the Sun and MD( „
8.23 Occurrence of Lunar Eclipses
Now that we have an expression for least angular separation of the Moon anrl
the antisolar point near opposition, let us examine how to use it to test for the
occurrence of a lunar eclipse. From the geometry of the shadow cone, we shall find
a relationship between the size of the cone and other well-known quantities.
8 231 ApparentSemidiametersof the Shadow Cones The penumbral shadow cone
of the Earth is defined by the interior tangents between the Sun and the Earth
S igure 8.231.1 illustrates the geometry. In the diagram, 5 is the center of the Sun E
is the center of the Earth, and M is a point on the penumbral cone at the distance
of the Moon. The line LM is the semidiarneter of the cone at the distance of the
Moon. The known quantities are angle GME, which is the lunar horizontal paral-
lax (nm) based on the mean radius of the Earth; angle SET, which is the apparent
semidiarneter of the Sun (*); and angle ETC, which is the parallax of the Sun (tts).
we wish to find LM as the angular quantity LEM seen from the Earth, designated
Figure 8.231.1
SSSC ZZT t ftSff ' PenUmbra' Shad°W> relatinS i,S ^ular to observed
quantities. In this diagram, ZGME = nm. ZETG = «•„ ZSET = js, and ZLEM = /, .
/ ECLIPSES OF THE SUN AND MOON
429
v
Figure 8.231.2
Geometric parameters of the Earth's umbral shadow, relating its angular size to observed
quantities. In this diagram, ZGME = 7rm, ZETG = tts, ZSET = i„ and ZVEM = f2.
fx. Angle EVM is an intermediate angle, so that angle EVM = angle ETC + angle
SET. We want /, = angle LEM = angle GME+ angle EVM. Thus,
/, = Km + ?rs +ss. (8.231-1)
The cross-section of the shadow is not a true circle, because the Earth is not
a true sphere. To compensate, a mean radius for the Earth is substituted, which
is equivalent to substituting for nm a parallax m reduced to latitude 45°, so that
fa = 0.9983407rm. Moreover, observation has shown that the atmosphere of the
Earth increases the apparent semidiarneter of the shadow by approximately one-
fiftieth. Hence,
ft - 1- 02(tti + tts + (8.231-2)
The umbral shadow cone of the Earth is defined by the exterior tangents be-
tween the Sun and the Earth. The vertex falls beyond the Earth and beyond the
Moon's distance as well. Figure 8.231.2 illustrates the geometry. In this diagram, S
and E represent the centers of the Sun and the Earth, respectively, and M is a point
on the umbral cone at the distance of the Moon. All angles are as described in the
Previous section, except that angle VEM, the apparent semidiarneter of the umbral
cone if 2), replaces angle From triangle VEM, f2 = angle GME - angle EVM, and
from triangle TEV, angle EVT = angle EVM = angle SET - angle ETG. Thus,
/2 = 1.02(7r1+7rs-Js) (8.231-3)
after compensation for the effect of the Earth's atmosphere on the semidiarneter of
the shadow.
8.232 Conditions for a Lunar Eclipse There are several types of lunar eclipse, de-
pending upon what portion of the Earth's shadow the Moon passes through. We
430 EXPLANATORY SUPPLEMENT
will consider the conditions for each. All conditions should be tested to find the
maximum possible eclipse.
8.2321 Conditions for a penumbra! lunar eclipse A penumbral lunar eclipse occurs
when the Moon enters partially or completely into the penumbral cone, but not the
umbral cone. The right side of Figure 8.2321.1 illustrates the limiting case for ;i
Figure 8.2321.1
Composite diagram showing limiting conditions for different types of lunar eclipses, not to scale.
Limiting conditions that separate types of eclipses occur when the separation of the center of the
Moon and the shadow axis is equal to the sum or difference of the apparent radius of the Moon and
the radius of the cross-section of one of the shadow cones at the distance of the Moon. Let the small
circle represent the Moon, with radius sm, the shaded circle represent either the umbra, with radius U<
or the penumbra, with radius f\ , and a represent the distance between centers S and M or M' . On the
left side, if a < f - sm, the Moon is totally immersed in the shadow. On the right side, if a > / + sm,
the moon is just out of the shadow. If the Moon lies between these two limits, it is partially eclipsed in
the shadow. The type of eclipse is named for its maximum phase — total (umbral), partial (umbral),
total penumbral, or partial penumbral.
penumbral lunar eclipse, where f - f\. In this diagram, S represents the antisolar
point and M' represents the center of the Moon. Let SM' be the geocentric least
angular separation (cr) given by Equation 8.22-9; M'T is sm. A penumbral eclipse
will occur if cr is less than f\ +sm. Thus, the condition for a penumbral lunar eclipse
is
cr < 1.02(tt, +tts +ss) + jm. (8.2321-D
8.233 Conditions for a partial lunar eclipse A partial lunar eclipse occurs when
the Moon passes through the penumbral cone and partly, but not completely, into
and through the umbra. The right side of Figure 8.2321.1 also illustrates the limiting
8 / ECLIPSES OF THE SUN AND MOON 431
case for a partial (umbral) lunar eclipse, except this time, / = fi. A partial eclipse
will occur if a is less than fa + sm. Thus, the condition for a partial lunar eclipse is
a < 1 . 02(7T! + tts - ss) + sm. (8.233-1)
8.2331 Conditions for a total lunar eclipse A total lunar eclipse occurs when the
Moon passes through the penumbral cone and completely into and through the
umbral cone. The left side of Figure 8.2321.1 illustrates the limiting case for a total
(umbral) lunar eclipse, where / = /2. In this diagram, S represents the antisolar
point and M represents the center of the Moon. Let SM be the geocentric least
angular separation (a) given by Equation 8.22 9; MT is sm. A total eclipse will
occur if a is less than fi~sm. Thus, the condition for a total lunar eclipse is
tr< L02(7n+7rs-ss)-jm. (8.2331-1)
8.24 Occurrence of Solar Eclipses
The geometry to test for the occurrence of a solar eclipse is similar to that for a
lunar eclipse. In this situation, we are examining the shadow of the Moon instead of
that of the Earth. The shadow is, of course, smaller and shorter, the vertex being at
about the same distance from the Moon as the Earth. Moreover, the Earth is much
larger, so that it can never be completely within the Moon's shadow. In searching
for the occurrence of a solar eclipse, we use the center of the Earth as the reference
point of observation. The expression for least angular separation of the centers of
the Sun and Moon still applies.
8.241 Conditions for a Solar Eclipse There are also several types of solar eclipse,
depending upon what portion of the shadow reaches the Earth's surface. We will
examine the conditions for each. It is not necessary to correct for atmospheric
augmentation of the Moon's shadow; the Moon has no atmosphere. It is also not
necessary to correct the parallax for shape; this is done by limb corrections, if
needed (Herald, 1983).
8-242 Conditions for a partial solar eclipse A partial solar eclipse occurs when
the Earth passes through the penumbra of the Moon's shadow, but not the umbra.
(Also, partial phase is seen by an observer in the penumbra during any type of
solar eclipse.) Figure 8.242.1 illustrates the limiting case for a partial solar eclipse.
In this figure, 5 represents the center of the Sun, M the center of the Moon,
and E the center of the Earth. Point O, on the Earth, lies on the edge of the
Penumbral cone, depicted by line segment OT, which also serves as the horizon at
432
EXPLANATORY SUPPLEMENT
Figure 8.242.1
Known geometric parameters in the limiting configuration for a partial solar eclipse, when the
Moon is externally tangent. As seen by an observer at O, the apparent disks of the Sun and the
Moon are externally tangent along the line OT. In this diagram, ZZOM = zL, ZZEM = zm
ZZOS = C ZZES = zt, ZMON = s'm, and ZSOT = si
O; point Z is at the zenith for an observer at O. The topocentric zenith distances
of the Moon and the Sun at O are:
z'm = 90° - ZMON = 90° - 4,
z's - 90° - ZTOS = 90° + 4. (8.242-1)
where s'm and s'g are the topocentric semidiameters of the Moon and the Sun, re-
spectively. The geocentric angular separation between the Moon and the Sun is
angle MES, which is equal to the difference between the geocentric zenith distances
of the Sun and the Moon, zs - zm- The geocentric zenith distances (z) are related
to their corresponding topocentric values (z1) by
Z = l! - ft', (8.242-2)
where -k is the geocentric parallax. Assuming a spherical Earth, 7r' is given in terms
of the horizontal parallax (w) by
tt' = sin 1 (sin tt sin z').
(8.242-3)
B / ECLIPSES OF THE SUN AND MOON 433
Substituting Equations 8.242-1 and 8.242-3 into Equation 8.242-2, neglecting the
augmentation of the semidiameters, and setting coss' equal to unity yields
Zm = 90 — Sm — 7Tm,
zs = 90°+ss-^s. (8.242-4)
Setting angle MES equal to the geocentric least angular separation (ct) yields
<T = Zs - Zm - + sm + ~ ^s-
(8.242-5)
A solar eclipse will occur when the apparent least angular separation of the centers
of the Sun and the Moon is less than the sum of their apparent semidiameters.
Thus, the condition for a partial solar eclipse as seen from the geocenter is
a < Ss +Sm +7Ym - 7TS.
(8.242-6)
8.2421 Conditions for a central solar eclipse If the observer on Earth passes
through the penumbra of the Moon's shadow and into the umbra, he or she expe-
riences a central solar eclipse. If the reference observing point is in the umbra on
the inside of the vertex, the eclipse is total. If the reference point is in the umbra
on the outside of the vertex, the eclipse is annular. If the reference point is in the
umbra and passes through the vertex, it is an annular-total eclipse. In the rare event
that an observer near a pole sees a total eclipse, but the shadow axis does not pass
through the surface of the Earth, the eclipse is a noncentral total eclipse. A central
solar eclipse will occur when the apparent separation of the centers of the Sun and
the Moon is less than the absolute value of the difference of the apparent semidi-
ameters. The condition for the occurrence of an annular solar eclipse is specifically
derived in this section, but the same condition may be used to differentiate between
the occurrence of any type of central solar eclipse (total, annular, or annular-total)
and the occurrence of a partial solar eclipse. Figure 8.2421.1 illustrates the geome-
try. The definitions in this figure are identical to the definitions of Figure 8.242.1,
except that point O now lies on the edge of the umbral cone. The topocentric zenith
distances at O are
4 = 90° + ZMON = 90* +C auttW)
z's = 90° + ZTOS =90° +4
Just as in the preceding section, the geocentric angular separation between the
Moon and the Sun is angle MES, which is equal to \Zs - Zml the difference between
the geocentric zenith distances of the Sun and the Moon. Employing the same
434
EXPLANATORY SUPPLEMENT
z
Figure 8.2421.1
Known geometric parameters in the limiting configuration for a partial solar eclipse, when the Moon
is internally tangent. As seen by an observer at O, the apparent disks of the Sun and the Moon are
internally tangent along the line OT. In this diagram, ZZOM = z'm, ZZEM = zm. ZZOS = z's,
ZZES = zs, ZMON = ?m, and ZSOT = s',.
procedure and approximations as in the preceding section, the condition for an
annular solar eclipse is
a < \ss - sm + ?rm - 7f8h (8.2421-2)
which, as stated before, may be used as a general condition for any central solar
eclipse. If ss > sm, the eclipse is total; if js < sm, it is annular. If a is near zero, the
eclipse may be annular-total, but that determination requires further calculation.
8.3 SOLAR ECLIPSES
Section 8.1 described considerations for ephemerides to use in calculating eclipses.
Section 8.2 described how to find approximately when an eclipse of any kind may
occur. This section describes how to calculate the details of a solar eclipse.
8.31 Fundamental Equations: Introduction
Bessel developed the method used to calculate and describe precisely any eclipse,
based on using a coordinate system oriented to the shadow axis. The basic steps are:
elements are calculated in the Besselian system to describe geometrical quantitites;
the observer's position is transformed to Bessel's coordinate system; equations of
condition are formed; circumstances are derived that describe the time and place
of observable events or conditions in the Besselian system; the circumstances are
transformed back to topocentric or geocentric coordinates.
ECLIPSES OF THE SUN AND MOON
435
NCP
Taxis by tt / 2 - d to make the new z-axis parallel to the shadow axis.
8.32 Besselian Elements
8.321 The Fundamental Coordinate System: Overview To use Bessel's method for
solar eclipses, the geometric position of the shadow of the Moon w.th respec o
the Earth is described in a coordinate system (Figure 8 321.1) ^re^on^
of coordinates is the geocenter, the z-axis is parallel to the axis of
positive toward the Moon, the x-axis passes through the equator and
toward the east, and the y-axis is positive toward the north. Thts ^£JX
tal coordinate system, and the ^-plane is the fundamental plan. The
of the shadow with the fundamental plane is two circles concentnc abou ttaax».
The surface of the Earth moves with respect to this plane, because the Earth
rotating on its axis and moving in orbit.
8.322 Calcn.ation of the Besselian E.ements: Transformation to the £
Coordinate System In the geocentric equatorial coordinate ^
position vector of the Sun, and Rm that of the Moon. Let G - Rs - Rm
collinear with the shadow axis. In astronomical units,
436
EXPLANATORY SUPPLEMENT
|RS| =
Is
rs =
rm =
Rs Rs sin 7rm '
G
horizontal parallax of the Sun at mean distance,
cos qs cos Ss
sin as cos <5S
sin<5s
cos am cos <5„
sin am cos <$„
sin 6m
sin7r0
Rs sin 7r„
By definition,
(8.322-1)
cos d cos a
g = rs - rm = g cos dsina , (8.322-2)
sind
where (a,d) is the geocentric equatorial right ascension and declination of the in-
tersection of the shadow's axis with the celestial sphere (point 2).
Recall from Section 8.321 the fundamental coordinate system (x,y,z) whose z-
axis is parallel to the shadow axis, or to g. The plane z = 0 is the fundamental plane,
the *-axis is the intersection of the fundamental plane with the equator (positive
to the east), and the y-axis completes the triad (positive to the north).
Hence,
rmFund. = ^r— - Rj (| - d) R3 (fl + |) rmGeoc, (8.322-3)
in units of earth radii. (R| and R3 are standard rotation matrices, positive direction,
about the jr-axis and z-axis, respectively.)
In the (x,y,z) system, let the unit vectors be i, j, and k. Then
g x k = 0,
g ■ i = 0,
g -j=o,
cos d cos a
cos d sin a , (8.322-4)
sin<i
k =
and the orthogonal unit vectors are
— sin a '
" — cos a sin d~
i =
cos a
j -
— sin a sin <i
. o J
cosd
(8.322-5)
/ ecL|psES OF THE SUN AND MOON 437
By working out Equation 8.322-3 or by resolving the lunar position vector into
conipoiients using Equation 8.322-5,
rmFund —
Sin 7Tm
cos 6m sin(ara - a)
sin 6m cos d — cos <5m sin d cos(am - a)
sin <5m sin d + cos <5m cos d cos(om - a) _
(8.322-6)
(X,y) are the coordinates of the intersection of the axis of the shadow with the
fundamental plane.
In presenting Besselian elements of eclipses for practical use, the right ascension
a of the point Z is conventionally replaced by the Greenwich hour angle (i of that
point, given by
fi - Greenwich apparent sidereal time - a.
In practical calculation, the Greenwich sidereal time is evaluated according to the
precepts given in Chapter 2.
8.323 Parameters of the Shadow Cone The radii of the umbra and penumbra on
the fundamental plane can be easily calculated given the generating angle of each
cone, and the perpendicular distance of each vertex from the fundamental plane.
Refer to Figures 8.323.1 and 8.323.2 for the plane geometry of a cross section of the
cone. This geometry relates the desired quantities to the known size and separation
of the bodies.
From Figure 8.323.1,
. , ds+da
sin/i
sin/2=^_^. (8.323-1)
gR*
To transform to units of Earth equatorial radii (see Figure 8.323.2),
de = 1 =Kssin7rs,
c;„ . (8.323-2)
sinss sinio
d, = if, sin ss - — = — ,
sin tts sin 7r0
where tts = 7To / Rs, ss = s0 / Rs- For the Moon, set
dm = k earth radii, where k-dm/de.
(8.323-3)
Fundamental plane "
/,
Figure 8.323.1
of ;shad°w cones ^ 'he Besselian fundamental reference system. The Sun is centered at
M with ™ h Semid'ameteur *- at heiSh' * above the fundamental plane. The Moon is centered at
Lush SSl £ at hf'!ht \ = z above the *""^amental plane. The fundamental plane,
enarftion hHno /tk ' ^ °bserver's Plane a«= perpendicular to the axis through SM, the
/ SXIh ™ iT , pem,mhra h3S i,S Vertex V> at heiSht f' above the fundamental plane, radius
Sub SnH rZ f P e> u 1 VhC 0bSerVer'S P'ane' Vert" an^le h ■ Similarly for the umbra.
tauDscnpt 1 refers to penumbra, 2 to umbra.)
/ ECLIPSES OF THE SUN AND MOON
Sun
439
Figure 8.323.2
Relationships among angular semidiameter s, apparent semidiameter d, and
parallax 7r. is expressed in Earth radii.
Then
de = 1 = Rm sin 7rn
dm=k = Rm sin sm =
sin sm
sin nm '
(8.323-4)
Let R = R. sin 7T(> be the distance to the Sun in astronomical units. Substituting
Equations 8.323-2 through 8.323-4 into Equation 8.323-1 and reducing yields
sin /] = (sin s() + k sin tt0) / gR,
sin fj = (sin s0 - k sin 7r0) / gR.
(8.323-5)
Thp numerators are evaluated using adopted values for k, sq, and 7rn. Equations
8.322-1 and 8.322 2 give g, and R is from the solar ephemeris.
The distances C\, C2, of the vertices of the penumbral and umbral cones above
the fundamental plane are shown in Figure 8.323.1. Figures 8.323.3 and 8.323.4
show that
c\ = z + /ccosec/i,
C2 = z — fccosec/i,
(8.323-6)
and hence the radii <i , I2 of the penumbra and umbra on the fundamental plane are
obtained from
/, =cj tan/,,
I2 = C2tan/2.
(8.323-7)
440
EXPLANATORY SUPPLEMENT
Figure 8.323.3
Vertex angle of the penumbra, detail from Figure 8.323.1, sin/, = dm I (c\ - z).
M
Figure 8.323.4
Vertex angle of the umbra, detail from Figure 8.323.1. sin/2 = dm I (z - c2).
The quantities z, k, c\, c2, h, and l2 are in units of Earth radii. The convention of
signs introduced in Equation 8.323-6 makes l2 negative for total eclipses— i.e., the
vertex is below the fundamental plane. For annular eclipses, l2 is positive, and of
course /[ is always positive.
8.324 Summary of Besselian Elements The quantities x, y, siad, cosrf, fi, h, fe>
lan/i, tan/2, A, and d are conventionally designated as Besselian elements. (N-B.
It is conventional in almanacs to use the notation (/>', d! for the derivatives with
respect to time.) In the publications, the first seven quantities are tabulated as a
function of time at a short interval, or may also be given as a low-order polynomial
as a function of time. The remaining four quantities, to the precision required, are
constant for the entire eclipse and are given at the conjunction value. Using all
these quantities, the size and orientation of the shadow cone with respect to the
8 / ECLIPSES OF THE SUN AND MOON 441
surface of the Earth and the fundamental plane may be calculated for any instant
during the eclipse.
8325 A Note on Practical Calculation In practice, using a large-memory
computer> one begins the computation by constructing short ephemerides of the
Sun and the Moon from the desired source, at a convenient interval, say, one hour,
and then subtabulates to smaller intervals as desired. However, it is possible with
some ephemerides stored in polynomial form to take the beginning ephemeris di-
rectly at the desired interval. This directness is especially helpful in calculating
points of curves near the beginning and end of a central solar eclipse.
Initially, for each eclipse, calculations are carried out using a ten-minute time
interval, but for some instances additional calculations at an interval of 30 seconds or
less are necessary. In all instances, the Besselian elements, derivatives, and auxiliary
quantities are calculated from the ephemeris entries and carried as an array; none is
assumed to be constant,. Derivatives are taken numerically. Most quantities change
so slowly that only a low-order scheme is required.
8.33 Coordinates of the Observer
8.331 Geocentric Position Consider an observer located on the surface of the Earth
in longitude A, at geocentric latitude 4>\ and at a distance p from the center of the
terrestrial spheroid. The longitude is not now corrected for AT and is considered
positive east (right-handed system). Careful attention must be paid to algebraic
signs (see Figure 8.331.1). In the y-, z-axes of the fundamental plane system, in
units of the Earth's equatorial radius, let the observer's position vector be
Pf =
(8.331-1)
In the geocentric equatorial system, let the observer's position be
COS 1
PC = P
b' cos A "
cos 4>' sin A
sin <b'
(8.331-2)
The transformation to the fundamental system is very similar to that of Equa-
tion 8.322-3:
PP -Ri (f - d) R3 (- (n - f )) Pc (8-331-3)
It is convenient to use the local hour angle in further calculations. Because by
current convention the Greenwich hour angle (p) and longitude (A) are measured
in opposite directions, the local hour angle (0) is
8 = V + X. (8.331^)
Applying Equation 8.331 3 to Equation 8.331-2, combining terms by multiple-angle
formulas, substituting Equation 8.331-4, and resolving the components along the
unit vectors gives
i = pcos 4>' sin 8,
V = psintp' cosd - pcos^'shWcosfl, (8.331-5)
C = psin 4>' s'md+ pcos 4> cos </ cos 0.
9 / ECLIPSES OF THE SUN AND MOON 443
8.332 Geocentric Velocity It will also be necessary to know how the observer is
moving with respect to the fundamental plane. The components of the motion are
rotations about the same two axes used in the coordinate transformation: rotation
about the polar axis of the Earth because of both the rotation of the Earth and
the motion of the Moon, and about the axis pointing to the equinox because of the
motion of the Moon. In the geocentric rectangular equatorial system, the rotation
vector is
Transforming to the fundamental plane,
(8.332-1)
wF = Ri {jj - d] u>G =
pcosd
{isind
(8.332-2)
The velocity of the observer with respect to the fundamental plane is
pF = wF x p¥ = R] ^ -dju>cx Pv
The operation o;p X may be expressed as a matrix premultiplier:
(8.332-3)
H = [d}x> U?y, U!z] —
Substitution of Equation 8.332-4 into Equation 8.332-3 yields
0 — psind pcosd
psind 0 -d
-pcosd d 0
(8.332-4)
Pf = ftp? =
p(-rjsind + (cosd)
p£ sin d — dC,
—pt£ cosd + dr)
(8.332-5)
8.333 Coordinates of Observer with Respect to the Fundamental Plane, Corrected
for Flattening Consider a spheroid of ellipticity e. Geodetic positions are specified
on the surface. Calculations of eclipse tracks in the fundamental plane are no prob-
lem, nor are local circumstances for a given point. However, in translating any
point from the fundamental plane to the surface of the spheroid, evaluation of C
requires knowing p, the distance of any point from the center of the Earth. This
distance is a function only of latitude, so that £ could be determined by successive
approximation.
However, Bessel devised a procedure to provide a direct computation of £ from
£1 V- This procedure requires a set of auxiliary Besselian elements, based on the
Besselian element d and the ellipticity e of the spheroid.
444
EXPLANATORY SUPPLEMENT
Let a and b be the equatorial and polar radius, respectively, of the spheroid
Then
flattening, / = - — -, and
a
ellipticity, e = (1 - b1 1 a2)112.
(8.333-1)
Define
C=(l-e1sin2<P)-m,
S = (l-e2)C;
or
C = (cos2 0 + (1 - f)2 sin2 4>rvl, S = (l _ ffc.
(8.333-2)
The relations between geocentric latitude <f>' and geocentric distance of observer a
and geodetic latitude 0 are
psin <f>' = Ssin<f>
pcos<j>' = Ccos0. (8,333-3)
To eliminate p and 0', transform to a parametric latitude 0i, defined by:
. . , cos <b
V»4n=P*»* =(1_e2sin2^}1/2"Ccos^ (8-333-4)
Then, from Equations 8.333-2, 8.333-3, and 8.333-4,
sin ^.j =(1 -cosVi)"^ =
(1 -e2)"2'
Inversely,
pcos 0' ~ cos if>\ =Ccos4>,
psin0' = (1 -e3)1/2sin<Ai sSsin0.
(8.333-5)
(8.333-6)
Substituting into Equation 8.331-5,
£ = cos <f>[ sin (9,
rj = sin 4> ] ( 1 - e2 ) 1/2 cos d - cos <f> i sin d cos 9,
C = sin0i(l - e2)1/2sinrf + cos0, cos d sin 0.
(8.333-7)
Now to eliminate the terms in d and e, define
8 / ECLIPSES OF THE SUN AND MOON
in 17:
445
in 0
Then
Define
sind - pi sind| ,
(1 - e2)1/2cosrf = pi cos^i,
(1 - <?2)1/2 sind = p2 sind2,
cost/ = p2 cosd2.
£ = cos0i sin#,
iji is fj j pi => sin 4>\ cost/i — cos <f>\ sin if 1 cos 8,
C - Msin 0i sin d2 + cos 4>\ cos ffc cos 0).
(8.333-8)
After extensive algebraic manipulation and reduction, this becomes
(8.333-9)
(8.333-10)
(8.333-1 1)
Q = sin 0i sint/j + cos0] cosd] cos#.
It is desirable to eliminate 0i and 6. Then Equation 8.333-9 combined with Equa-
tion 8.333-11 may be written
" cos 0i sin (9 "
m
sin 0i
A.
cos 0| cos 9 _
Hence,
' cos 0i sin# "
T
sin 0i
= Ri(rfi)
m
cos 0i cos 9 _
(8.333-12)
(8.333-13)
Apply components of Equation 8.333-13 to the expression for C in Equation 8.333 9
and reduce to obtain
C = p2[G cosCrfj - d2) - m sinWi - d2)l (8.333-14)
8.334 Auxiliary Besselian Elements and Summary The following defined quanti-
ties are part of a set known as auxiliary dements; more are defined in Section 8.3422.
These elements are used to introduce the flattening of the Earth into the eclipse
calculations. All quantities are in units of Earth's equatorial radius.
446
EXPLANATORY SUPPLEMENT
pi s(l -e2cos2d)m,
P2 = (1 -e2 sin2 d)ia,
sin^i = sin J/ pi,
cos rf| = ( 1 - e2)"2 cos d I p\ ,
sin(c?i - (I2) = e2 sinrfcosc// pyp2,
COS(rf, -<fe)a(] -e2)1/2/plP2.
(8.334-1)
(The subscripts do noi refer to the umbra and penumbra.) In computation, all these
quantities are carried with the regular elements at the same interval, even though
pi, P2, and (d\ - di) are nearly constant for the duration of the eclipse. Once the
elements are available and a point (£, T),0) is given in the fundamental plane, then
4> and Q are obtained as follows: point (£, t?i,Ci) is calculated from the definitions
of ?7i and Ci in Equations 8.333-9 and 8.333-10. Next, Equation 8.333-13 gives fa
and 9; Equation 8.333-6 gives fa and Equation 8.333-14 or 8.333-7 gives £.
8.34 Conditional and Variational Equations
8.341 Introduction This section establishes conditions by which eclipse phenom-
ena may be distinguished and calculated.
8.342 Conditions Defining General Circumstances and Curves Consider now the
relation between observer and shadow. We have, in the fundamental plane reference
system, the geocentric position of the observer p = (£, 77, 0. The geocentric position
of the intersection of the shadow axis with a plane through the observer parallel to
the fundamental plane is
Fx~
r -
(8.342-1)
Now define the vector in the observer's plane from observer to shadow axis as
A = r - p. (8.342-2)
In the plane though the observer, let Q be the position angle of the axis, measured
from north through east, so that the unit vector
A =
sing
cos Q
0
(8.342-3)
The radius (L) of the shadow at height f above the fundamental plane is
L = /-Oan/,
(8.342-4)
8 / ECLIPSES OF THE SUN AND MOON 447
subscripted for umbra or penumbra, and L may be positive or negative.
g 3421 Equations of condition We may now impose geometric and dynamic con-
ditions in order to calculate eclipse phenomena. We know r. We may consider two
cases from Equation 8.342-2: one, in which p is not known but we wish to find the
locus of all points satisfying a condition as a function of time (curve); the other, in
which p is fixed and we want the times of significant configurations (circumstances).
Much of the following discussion applies to either case, the difference being the
choice of knowns and unknowns. First, consider the more general cases, in which p is
not known. The discussion applies to both umbra and penumbra, so the subscripts
are omitted.
If the observer is on the edge of the shadow (penumbra or umbra) or, more
generally, the surface of the shadow cone, the eclipse is either beginning or ending.
This surface location may be defined by imposing the geometric condition
\A\ = \L\, or A-A-L2 = 0. (8.3421-1)
Consider Equation 8.3421-1 now as a general expression that can be evaluated at
any instant. The equation has two roots. If they are imaginary, then the observer is
outside the shadow. If they are equal, then the eclipse is simultaneously beginning
and ending i.e., "grazing"— and the observer is on either the northern or southern
limit of the shadow. This limit may be defined by imposing the dynamic condition
—(A ■ A - L2) = 0,
at
or
A-A-ZX = 0. (8.3421-2)
Since
|A| = |L|, this reduces to A • A - L = 0. (8.3421-3)
It is always the case that df I dt = 0 for /, and f2, so it is sufficiently accurate to use
L = /-Ctan/,
In Equation 8.3421-3, substitute
A = r - p,
p=WFXpF = wFx(r- A),
448
to get
A ■ r - A ■ (wF x (r - A)) + tan/wz • (r • A) - I = 0
In the second term,
EXPLANATORY SUPPLEMENT
(8.3421-5)
A • (wF x (r - A)) = A • (a>F x r).
(8.3421-6)
From Equation 8.332-4:
" 0
-fismd
ficosd"
~ x~
us x r =
fi sin d
0
-d
y
—fi cos d
d
0
X.
= V-
— y sin d+ (,cosd
x sin d — C,d I fi
_ -xcosd + ydl fi
" —y sin d '
—ficosd
xsv&d
-c
d
_ —xcosd_
. -ydK .
(8.3421-7)
From Equations 8.332-2, 8.332-4, and 8.3421-1:
" d '
ficosd'
Wp -
ficosd
0
_ ft sin d _
0
A = A(/-Ctan/).
(8.3421-8)
Substitute these and Equations 8.3421-1, 8.3421-6, 8.3421-7 into Equation 8.3421-5
and, through algebraic manipulation, collect terms as factors of A- and ( to obtain
" — y sin d '
x sin d
- /tan/w, j
—xcosd^
- 1+ tan M ■ r = 0. (8.3421-9)
Equation 8.3421-9 may be written in component form as:
sin Q
cos Q
0
+ Csec2/
x
y
LC.
sin 2
cos g
0
ys\nd
—x sin
jccos<i
■ficosd
d
0
/tan/
/ + tan /
—ficosd
d
0
—ficosd
d
0
(8.3421-10)
T r I
X
y
= 0.
" sin Q
T
C
' sin Q '
r
" —ficosd
cos<2
-b
+ ( sec2 /
cos Q
d
0
. 0 _
0
0
a / ECLIPSES OF THE SUN AND MOON 449
g 3422 Additional auxiliary elements With a large computer, Equation 8.3421-10
niay be used in this form to find combinations of (, Q that satisfy it, and hence
the limiting curves. For efficiency, however, it is conventional to define auxiliary
quantities a, b, c with derivatives a, b, c, there being a separate set for umbra and
penumbra as distinguished by subscripting and use of the corresponding angle /,
such that
+ d = Q. (8.3422-1)
Thus, to make Equation 8.3421-10 identical to Equation 8.3422-1,
a = —I — fvccosdtanf + _yc?tan/,
b = -y + fvcs\nd + Wtan/, (8.3422-2)
c-k +jj,ysmd + 1 ft tan / cos d.
In older references, the approximations d tan ft = 0, sec2 /,■ = 1 are used, which are
reasonable and adequate for limited computer resources.
Now collect terms in sin Q, cos Q:
a + sin Q(c - fi( cos d sec2 f) + cos Q(-b + Qd sec2 f) = 0. (8.3422-3)
Multiply by sec Q to obtain
c — fi\ sec J cos d
The application of this fundamental equation to calculate limits of the shadows will
be described in Section 8.355. In calculations for The Astronomical Almanac, the
auxiliary quantities are calculated along with the Besselian elements and carried in
an array as a function of time.
8.343 A Note on Practical Calculation The Besselian elements, both regular and
auxiliary, are nearly linear in behavior during an eclipse, such that integration
or numerical differentiation seldom requires more than a third-order Lagrangian
scheme, and linear interpolation is often adequate. In preparing computer programs
and carrying elements in an array, however, it is always necessary to be sure that
angular quantities change monotonically, especially when passing through 360° . It
is also necessary to carry enough entries at. the beginning and end of the eclipse to
allow interpolation, but not so many that the behavior of the tabulated elements
accelerates or reverses and potentially creates specious extrema.
450 EXPLANATORY SUPPLEMENT
Even though in calculations for The Astronomical Almanac all quantities are
carried in double precision (15-16 digits), the published data should be considered
accurate only to the number of significant figures presented.
8.35 Calculation of General Solar Eclipse Phenomena
8.351 Introduction The Besselian and auxiliary elements are the basis of calcu-
lating all further phenomena associated with each eclipse, utilizing, of course, the
necessary ephemeris data given in the eclipse section or elsewhere in The Astronom-
ical Almanac. The Eclipse section also gives positions on the surface of the Earth
as a function of time, where well-defined phases or phenomena occur according to
certain geometric or dynamical conditions— e.g., geocentric beginning and end of
each phase. These data are given in tabular and graphical form, and are known
as general circumstances, elements of the eclipse, and curves of several kinds. The
independent variable, unless stated otherwise, is Universal Time. Conversely, for a
given imposed condition, a time may be either deduced or assumed, and the coordi-
nates (£, n,Q) determined. Using the auxiliary elements pi, P2, d\, d2, the quantities
C and (f,77i,Ci) may be calculated and then transformed to A, &, 4>i,4>- General cir-
cumstances are usually calculated before the curves, but for purposes of exposition,
calculation of the curves is described first.
Eclipse curves are the loci of points on the spheroid of the Earth at some
specified radius, usually the surface. These curves delineate set boundary conditions,
and hence define areas within which certain eclipse phenomena are visible; they also
show time of occurrence. They are produced by going through a table of Besselian
elements as a function of time, calculating a table of derived discriminants, and
evaluating whether imposed criteria are satisfied.
General circumstances (as opposed to local circumstances) are individual points
and times at which extrema of the curves occur on the spheroid of reference.
8.352 Basic Geometry In general, eclipse phenomena are treated and calculated
using the fundamental triangle in the fundamental plane shown in Figure 8.352.1.
By imposing various conditions and finding corresponding points in the funda-
mental plane, it is then but a relatively easy step to project the point to the surface
and convert to latitude and longitude as described in Section 8.33.
8.353 Relationships The fundamental relationship of the geocenter (0,0), the pro-
jected axis of the shadow (x,y), and the projected position of the observer (6 *7) 1S'
p = m-A. (8.353-D
j ECLIPSES OF THE SUN AND MOON
451
Figure 8.352.1
Relationship between rectangular and polar reference systems with the origins at the
geocenter or the observer's projected point, in the fundamental reference plane. Angles are
measured from the north, or y-axis. In the geocentric system, the observer's projected point
may be specified as (£, n) or (p, 7); the shadow axis intercept may be specified as (x, y) or
(m,M). In the reference system centered at the observer's projected point, the position of the
shadow axis intercept is (x — £,y — n) or (A, Q).
! derived in Section 8.3421, if the point of the observer is "tangent" to the path of
the shadow, such that the eclipse is beginning and ending simultaneously, then the
second fundamental relation, which is a condition on Q, is (from Equation 8.3422-3)
a, - bi cos Q + a sin Q + C( 1 + tan2 /,-)(<* cos Q - p cos d sin 0 = 0, i = 1 or 2.
(8.353-2)
P is usually not unity; it contains flattening and height above the geoid.
These relationships are also used:
' x - A sin Q
psin 7
_v _
y — A cos Q
pcos 7
(8.353-3)
452 EXPLANATORY SUPPLEMENT
Furthermore, the cosine law for plane triangles (dot-product) gives
A A = (m - p) (m - p) <=> cos(7 - Af) = - ,
Imp
m2 + A~ — o2
p ■ p = (m - A) ■ (m - A) <=> cos(0 — M) = - — ^ , (8.353-4)
2m zi
m ■ m = (p + A) ■ (p + A) <=> sin(7 - 0) = - .
2Ap
The sine law for planar trigonometry (cross-product) gives
m x A = kZi(jccos Q-y sin Q)
= kAm $in(Q - M),
pxA = (m-A)xA = mxA
= kpzlsin(7 - 0. (8.353-5)
Another useful relation is
„ . i 1 . _ , . .„ (I — m + p)(l + m — p)
2 sin- -(7 - M) = 1 - cos(7 - jW) = - ^ ^. (8.353-6)
2 2wp
8.354 Flattening Flattening is taken into account by altering p and 7 slightly to
rp and 7'. Assume that £2 + ??2 + C? = r\ Now if Ci = 0, then £2 + ?>2 = 12. Let
£, = rpsin7 = rsin7',
^ = rp cos 7 = rp] cos 7', (8.354-1)
771 = 77 / pt - /-COS7'.
Then
tan7' = ^/ni = p,f /t? = pi tan 7, (8.354-2)
and
sin 7' cos 7' ,0 **ax\
— =p = p} . (8.354-J)
sin 7 cos 7
The following iterative procedure accounts for flattening. Ordinarily, three itera-
tions are sufficient, for each of ± (7 - m).
To start, assume |p| = ], where p = |f, r)\ from Equation 8.353-3 and Fig-
ure 8.352.1
(a) sin 7 = £ / p, cos 7 = n I p;
g / ECLIPSES OF THE SUN AND MOON 453
(b) tan 7' = p[ sin 7 /cos 7;
(c) 7' = tan" '(tan 7');
(d) p = sin 7' /sin 7. (8.354-4)
Repeat the steps of Equation 8.354-4 three or more times, until convergence.
Then £ = sin 7', -rj\ = cos 7' (and Ci =0).
8.355 Curves The order in which curves are computed is not important: compu-
tations are all done independently from the tabulated elements. However, it helps
to leave computation of the outline curves until last.
8.3551 Rising and setting curves An observer who is at a point of intersection of
the penumbral cone with the fundamental plane is also on the sunrise/sunset ter-
minator and may, weather permitting, see partial eclipse beginning or ending at
sunrise or sunset. The locus of all such points on the Earth's surface, known as the
rising and setting curve, is shown graphically on eclipse maps in The Astronomical
Almanac with the label Eclipse begins at sunrise, Eclipse ends at sunset, and so on.
The rising and setting curve is the locus of end points of the outline curves for
the penumbra as it intersects the terminator. At a selected instant, there are two
points on the curve or no points. If there are two points, each belongs to one of two
series. For a succession of instants, the two series of points trace different branches
of the same curve and are joined at first contact (penumbra emerges from the
terminator and intersects the Earth's surface) and last contact (penumbra merges
completely into the terminator and leaves the Earth's surface). The track of the
penumbra always generates at least one limit curve (Section 8.3553), either northern
or southern, but if both the northern and southern limits exist for an interval of t ime
(i.e., the central path is in equatorial regions), then during that interval the rising
and setting curve does not exist — the penumbral shadow is completely emerged
from the terminator. The result is that the rising and setting curve forms two
separate loops like elongated teardrops, each loop being generated in two segments.
Otherwise, if there is only one limiting curve, then the loops are joined, and the
curve of rising and setting forms a distorted figure-eight. In either case, one loop
or lobe is for rising, the other for setting. In a series of similar eclipses, both forms
occur. When the first separation of the loops occurs in the series, the break into
separate loops does not occur at the apparent node, but near it.
In the fundamental plane, the circle of the penumbral cone, at any instant,
intersects the surface of the Earth in no points or in two points (which may occa-
sionally be coincident). In the fundamental triangle,
£ = 0, A - l[ are imposed conditions.
(8.3551-1)
454 EXPLANATORY SUPPLEMENT
If Rj is the matrix for positive rotation about the z-axis, then
p=-R?(+(j-Af))m.
m
(8.3551-2)
R. =
cos(± (7 - AO)
^ sin(± (7 - AO)
0
±sin(±(7-A0) 0
cos(+ (7 - AO) 0
0 1
(8.3551-3)
where one uses all top signs or all bottom signs.
If, by the cosine law (Equation 8.353-4), cos2[± (7 - m)] < 0, then there are no
points on the curve for the selected instant.
8.3552 Curves of maximum eclipse in the horizon Any locus of points that satisfies
the fundamental relation given in Equation 8.353-2 (a dynamic constraint) is a
curve of maximum eclipse. If the geometric conditions are also imposed that £ = 0,
A = l\, then we have the curve of maximum eclipse at sunrise or sunset, or maximum
eclipse in the horizon. On the map this curve approximately bisects the lobes of
the rising and setting curves. It is continuous if the rising and setting curves are
joined, otherwise not. It is also generated in two branches, as are the rising and
setting curves.
From Equation 8.353-2,
< = u
a 1 - b\ cos Q + cp sing = 0> (8.3552-1)
and from Equation 8.353-5,
pxA = mxA=> psin(7 - Q) =xcos Q - y sing. (8.3552-2)
From Equation 8.3552-1 find two values of Q, then for each find a value of 7, then
7' as before (Equation 8.354-4).
Points exist on these curves only if both these conditions are satisfied:
sin(7-0 < 1,
(x-02 + (y-ViPif <l\-
(8.3552-3)
(8.3552-4)
8 / ECLIPSES OF THE SUN AND MOON 455
8.3553 Central line, duration of central eclipse, and width of path The cent ral line
is the locus of the points of intersection of the axis of the shadow with the surface
of the Earth. The following conditions must be satisfied:
p ■ m = pm,
pxm-0,
? =x, 7) = y, 171 =yl p\,
C2 + ^ + Cf = ^. (8.3553-1)
From Equations 8.3553-1 and 8.333-14:
Ci = +</*-
C = cosfrfi - d2) - Vi m(di - dz)l (8.3553-2)
Points on the opposite side of the spheroid may be found by taking the negative
values of (j-
The duration of central eclipse may be estimated by disregarding vertical mo-
tion and considering only the instantaneous size and velocity of the intersection of
the shadow with the plane through the observing point and parallel to the funda-
mental plane.
From the section on conditional and variational equations (Equation 8.3421-5):
pp = UJF X pp
has components
4 = p(—y sin d + C cos d),
17 = fix sin d - dQ. (8.3553-3)
Using the definitions
n2 = (x - if + (y - f,)2,
L2 = h - Ctan/2,
the duration is
2L2/n (with appropriate sign). (8.3553^4)
The units must be converted as necessary. Normally the conversion is from hours
to seconds of time. Before 1982, L2 had to be adjusted for different values of k.
In local circumstances, however, the accurate duration is the difference between
Appropriate pairs of contact times.
456 EXPLANATORY SUPPLEMENT
The width of the path, perpendicular to the direction of motion, may be esti-
mated by a formula derived by Mikhailov (1931):
width = 2L2 < C2 +
-i 5 '
1/2
n n
(8.3553-5)
8.3554 Northern and southern limits of umbra or penumbra Each curve limiting
a shadow path is the locus of points on the surface of the Earth's spheroid tangent
to the path of the shadow, such that first and last contacts are simultaneous, or else
maximum occurs in the horizon. The umbra nearly always has both limits on the
surface because it is so small, but the penumbra often has only one. The northern
and southern limits of the umbral track define the central path; the northern and/or
southern limits of the penumbral track, plus the curve of rise and set, define the
limits of the eclipse.
Impose the condition that Equation 8.353-2 must be satisfied. Choose a value
of £. If a complete set of limits is being generated, it is usual to begin with £ = 0.
Then the coefficients of sin Q and cos Q are evaluated and the left-hand side of
Equation 8.353-2 is used as a discriminant. This discriminant is evaluated for a
run of Q, starting from a first guess obtained by setting a, = 0, and proceeding a
degree at a time until two zero points are found (by inverse interpolation when the
discriminant changes sign). Then for each Q, iterate on C as follows, using Besselian
elements:
Li = {; - Ctan/,-,
£ =x — Lj sin Q,
??i =(y - L,sin(2)/ pu
C? = i2 - £2 - t}\ (If < 0, there exists no point.)
Normally choose Q > 0 to continue.
Ci < 0 may be used to continue the curve below the horizon.
C = PilCi cos(rf, -d2)-vi sin(</, - d2)l (8.3554-1 )
i = umbra or penumbra.
Put this value of C into 8.353-2, and repeat until it converges to the required
tolerance. In The Astronomical Alw.an.ac, that is 10-5 earth radii. In extremes of
these curves, the iteration may begin oscillating about some mean value, in which
case the mean is taken to continue the iteration.
When the iteration converges, then the point is assigned to a limit, by the
discriminant:
Lj cos Q < 0 for northern limit
LjCosQ > 0 for southern limit. (8.3554-2)
8 / ECLIPSES OF THE SUN AND MOON
The coordinates (£, ni.Ci) are then easily converted to (4>, A).
457
8.3555 Outline curves An outline curve of the shadow (umbra or penumbra) on
the spheroid is the locus of all points at which the corresponding central or partial
eclipse is beginning or ending at a given time. Outline curves are usually published
only in the form of special maps. Outline curves for the penumbra, showing the
places where the partial phase is beginning or ending at stated times, are plotted
on the small-scale maps presented in The Astronomical Almanac and other publica-
tions. (See also Section 8.361.) Outline curves of the umbra are not published. For
all practical purposes, it is acceptable to neglect the flattening of the Earth in these
calculations, but for the maps in The Astronomical Almanac, it is not neglected.
The angle Q becomes the independent variable. At the instant for which the
curve is calculated, if both the northern and southern limits of the penumbra exist,
then Q has the complete range of 0° to 360°. If not, the extreme values must be
found; they are two points on the curve for which ( = 0. Because these points are
on the limiting curves, in the computer program an initial test is made for the
availability of points already computed for the limiting curves.
Those points are then verified from the equations
£ - x - Li sin <?,
niPi -y- Li cos Q,
e2+^ + Cf = 1, (8.3555-1)
in which £i = 0. Once the end points are found and the correct portion of the
arc which intersects the illuminated surface is established, points are calculated at
one-degree steps of Q, iterating as described for the northern/southern limit curves
and requiring (i > 0. If desired for special purposes, points below the horizon may
be found by using the solutions for Ci < 0.
8.3556 Other curves For 1960-1980, maps in the American Epkemeris and Nau-
tical Almanac showed curves of equal middle and equal semiduration of eclipse.
These curves were constructed graphically from the intersections of a network of
outline curves, as there is no way to calculate them directly.
Curves of maximum eclipse are not prepared for regular publications, but may
be readily calculated. Such calculations are an extension of the procedure for the
curve of maximum eclipse in the horizon, in which the condition £ = 0 is removed,
and £ becomes the independent variable — i.e.,
Choose C;
Find Q (two values) from Equation 8.353-2;
Find 7 from sin(-y -Q) = (xcos Q - y sin Q) I rp (fundamental cross-product).
458
EXPLANATORY SUPPLEMENT
s X n
Figure 8.3556.1
Definition of auxiliary quantity ip. L is radius of the shadow in the plane of the observer,
A is distance of observer from the axis, n is speed of separation, s is semi-duration.
If neither value of Q gave an imaginary result, only one may be an eclipsed point
in most cases. It is also required that
\A\ < \L\, or (jc - 01 + (y ~ "H? > if. (8.3556-1)
If this test is met, iteration may proceed as usual.
From maximum eclipse, semiduration of the partial phase may be estimated as
follows. Define tp in Figure 8.3556.1, where s is the semiduration and n is the speed
of shadow. The respective units are hours and equatorial radii per hour. Then
s = (Lj cos ip) I n,
where
smip = A I Lj,
n2 = (x-02 + (y-r))\
I = fi cos ip cos 6,
V = Ksind. (8.3556-2)
The exact duration is found from the difference in times of first and last contacts,
as described in Section 8.3621.
The magnitude of the eclipse is found from
k—A. (8.3556-3)
L, +L2
If Lq. is not available, the approximation La = £-i — 0.5459 is used. (Subscript 1 for
penumbra, 2 for umbra.)
g / ECLIPSES OF THE SUN AND MOON 459
Curves of equal semiduration or equal magnitude must be found by inverse in-
terpolation on curves of maximum eclipse. (See Section 8.36 on local circumstances
for further discussion of magnitude.)
8.356 General Circumstances A collection of event points known as the general
circumstances are the important extrema of the curves of the eclipse. They define
the bounds of the eclipse on the spheroidal surface and in time. They also join
together the various curves and are essential in preparing the eclipse map. General
circumstances are computed from the Besselian elements through a set of discrimi-
nants. For each discriminant, a table of values is calculated for a sequence of times,
and used as described in Sections 8.3561-8.3565. In practice, general circumstances
are calculated before the complete curves are generated. If circumstances are re-
quired for high altitudes, the calculations are repeated for a larger spheroid.
The first and last contacts of the penumbra occur for every solar eclipse. They
are the points where the shadow cone of the Moon first and last encounters the
spheroidal surface. These points are shown on the eclipse maps as First Contact
and Last Contact, and always lie on the rising and setting curve. The times at which
they occur define the date(s) of the eclipse.
If the eclipse is central, then there are first and last contacts of the umbra. These
are the extreme end points of the central line, and fall on the curve of maximum
eclipse in the horizon.
For a central eclipse, there are also extreme points of the northern and southern
limits of the central track. It is a rare occurrence when only one limit exists. These
points also fall on the curve of maximum eclipse in the horizon.
The extreme points of the northern and southern limits of the penumbra fall on
both the rising and setting curve and the curve of maximum eclipse in the horizon
and, thus, mark the joining points. The extremes need not be known accurately;
they are the points at which an observer theoretically could "see" the limb of the
Moon brush the limb of the Sun for an instant. If the eclipse curves have been
calculated taking flattening of the spheroid into account, the same must be done
for these points or the curves will not fink.
The point of central eclipse at local apparent noon or midnight is self-explanatory,
i.e., maximum occurs on the meridian. Meridian passage at midnight may occur m
the polar regions.
The point of greatest eclipse is calculated only for partial eclipses and shown
on the eclipse map. It is the point on the spheroidal surface that comes closest to
the axis of the shadow, and it lies on the curve of maximum eclipse in the horizon.
8.3561 First and last contacts of the penumbra First and last contacts of the penum-
bra occur when the shadow cone is tangential to the spheroid. Thus at two instants,
x2+y2 = (/, +P)2.
(S.3561-1)
EXPLANATORY SUPPLEMENT
460
Taking the flattening of the Earth into account, let
2 2 7
m =3T +y%
m\ = JT+yr,
yt = yi p,
P = ™l>n\- (8.3561-2)
Then x2 + yj = (/, + m / /n, f is the condition; the discriminant is
Di s.r2 + vf - (/[ +mlm\)1. (8.3561-3)
In the table of discriminants, the times of the contacts are calculated by inverse
interpolation when D, = 0. These times are used to obtain*, y, and Pl. At the times
of contact,
m =y\ i m = v/ p\m\,
0=° (8.3561-4)
gives the data to compute latitude and longitude by successive approximation.
8.3562 Beginning and end of central eclipse Central eclipse begins and ends when
the shadow axis is tangential to the Earth's surface— i.e.,
0=0. (8.3562-1)
The discriminant is
D2=x1+y]-\, (8.3562-2)
and the procedure is the same as for the penumbra.
8.3563 Extreme points of umbral and penumbral limits The extreme points of the
limits of umbra and penumbra are found by setting Ci = 0 and then calculating tan Q
from Equation 8.3422-4.
cos0 = ±(l +tan20-|/2,
sin £ = cos Q tan Q,
8 / ECLIPSES OF THE SUN AND MOON 4g-|
% = x - li sin Q, (=1,2,
rj) = (yT//cos0/pi, (= 1, 2,
f2 + 'fr = l. (8.3563-1)
Note that the sign of cos Q determines whether the point is on the northern or
southern limit. Note also that there are two cones and, therefore, four discriminants
of the form
m*(x-lsmQ?+ (y~^°SQ) -1, i = 3,... ,6. (8.3563-2)
Again, the times are found by inverse interpolation, and the coordinates, with
flattening taken into account, by the usual procedure.
8.3564 Central eclipse on the meridian Central eclipse on the meridian — noon or
midnight —is the point on the central line at conjunction of the Sun and the Moon
in right ascension. The discriminant is x, and the condition is x = 0. Test the signs
carefully in polar regions, to distinguish noon from midnight.
8.3565 Greatest eclipse In partial eclipses, the maximum magnitude, or greatest
eclipse, occurs at the point on the surface of the spheroid that conies closest to the
axis of the shadow. The magnitude changes very slowly, so high precision is not
required. The eclipse will occur in the horizon at this point, and it is sufficient to
find the time when the shadow axis is closest to the center of the Earth or when
the rate of change of separation goes to zero. Hence, the discriminant is
D-,sxx + yy. (8.3565-1)
When the time is known, position is found as before. The distance of the observer
from the axis is
A = m - p. (8.3565-2)
he magnitude of the greatest eclipse for this case is
(8.3565-3)
l] + 12
8.36 Local Circumstances
Before proceeding, the reader should review Section 8.34 on conditional and vari-
ational equations. Local circumstances provide a description of eclipse phenomena
relative to a fixed reference position, which is equivalent to fixing p as constant.
462
EXPLANATORY SUPPLEMENT
8.361 Eclipse Maps The eclipse maps currently given in The Astronomical Al-
manac show the region over which various phases of each solar eclipse may be seen
and the times at which phases occur. Each map is a plot of the curves described
in Section 8.355, except that outline curves are limited to those of the penumbra
every half hour.
The outline curves are divided into leading edge (in short dash) and trailing
edge (in long dash). Except for certain extreme cases, the shadow outline moves
generally from west to east. For a given location, first contact (the beginning of
partial eclipse) occurs when the leading edge of the shadow arrives; similarly, last
contact (the end of partial eclipse) occurs when the trailing edge arrives. First or
last contact may be estimated from the map to within a few minutes.
8.362 Precise Calculations There are many refinements to be considered in per-
forming precise calculations: choosing a consistent ephemeris, corrections for Earth's
rate of rotation, and, for the Moon, offset of center of figure from center of mass;
correcting for irregularities of the lunar limb; and effects of the Earth's atmosphere
(refraction in solar eclipses and the effect on the shadow in lunar eclipses).
Some elements (e.g., the radius of the shadow cone) used to describe an eclipse
remain nearly constant for the duration. Elements that change do so very nearly
linearly. In effect, an observer sees an eclipse as two disks of fixed size, one crossing
the other in a straight line at constant speed.
In conventional notation, define u, v such that:
r- p =
y-rj
V
0
0
The distance from the observer to the nearest point of the shadow axis is |m| = m.
Let the relative velocity be n = m, and then the relative speed is |n| = n. The rate
of separation is m • n, and maximum eclipse occurs when the shadow axis is closest
to the observer, or m ■ n = 0.
8.3621 Contact times and duration At the beginning or end of penumbral, or par-
tial, phase, the condition b«a£j. Similarly, if the chosen point is in the central
path, then at the beginning or end of umbral phase m = La-
There are methods for finding the times of these phenomena by successive iter-
ation, using approximations and auxiliary angles. However, it is simpler to tabulate
h, v, L (where L is either radius) as a function of time, and also the discriminant
u +v2 —L2 . When the discriminant goes to zero, inverse interpolation gives the time
of the local contact. Duration is the difference between the two contact times for
the umbra.
g / ECLIPSES OF THE SUN AND MOON 463
8.3622 Position angles The projected shadow is a reflection of what the observer
sees in the sky, so the position angle of contact, Q, may be found from
m =
L&inQ
L cos Q
0
(8.3622-1)
The angle Q is measured eastward from the north, on the solar limb as the observer
sees it, or the projected shadow circle on the imaginary plane of reference. From
the observer's point of view, however, the vertex (up) is not usually north. The
parallactic angle C, defined by tanC s is the measure of the difference. The
position angle V of the contact point from the vertex is
V = Q - C.
(8.3622-2)
8.3623 Magnitude and obscuration Magnitude is commonly confused with obscu-
ration. Magnitude is defined as the fraction of the solar diameter covered by the
Moon at the time of greatest phase and is expressed in units of the solar diameter.
Obscuration is the fraction of the surface of the solar disk obscured by the Moon.
In the case of total eclipse, the magnitude can be greater than 1.0. To derive
the expression, however, the annular case is shown in Figure 8.3623.1. An observer
at point B in the penumbra sees part S'B' of the solar diameter blocked off, whereas
an observer at point E in the umbra sees the Moon's entire disk projected in DD'
on the Sun, SS'.
The known quantities are
P'O = PO = Lu
A'0 = AO = L2,
BO = m.
(8.3623-1)
At B, magnitude M\ - S'B' I SS'
= PB I PA' by simple proportion
= (PO-BO)/(PO+OA')
p (L] -m)/(L, +L2).
(8.3623-2)
At E, magnitude M2 = DD' I S'S
= (S'D - S'D') I S'S
= PE/PA' -AEIPA'
= PA I PA'
= (JU -L2)/(L, +La).
(8.3623-3)
Figure 8.3623.1
Magnitude is the fraction of the linear diameter of the Sun covered by the Moon. In this schematic
of an annular eclipse, in the observer's plane, A' A is the zone of annularity. To an observer within
■ n'n1Cc*at P°lnt £' the M°°n'S cross"section MM' Projects onto SS' as DD'. Hence the magnitude
is U U I S S. To an observer at point B in the zone of partial eclipse, S'B' is obscured, hence the
magnitude is S B I SS.
Figure 8.3623.2 , , , , . .
Obscuration is the fraction of the area of the solar disk obscured by the Moon.
Here it is the ratio of the area of segment CDC'F to the area of the solar disk.
For a total eclipse, in which the plane is on the other side of the vertex, A and A
are reversed, so that OA = -L2, and the expression is identical. Note also that this
is the diameter of the lunar disk in units of the diameter of the solar disk.
The degree of obscuration is now calculated using Lu U, and m- In
ure 8 3623 2 A is the center of the lunar disk, B is the center of the solar disk.
The disks overlap, with intersections at C, C . The line AB joins the centers and in-
tersects the circumferences and the chord CC at the indicated points. Let the^ lunai
radius be s = AC. The solar radius is taken as unity; hence BC = 1 and AC - M2 - S,
By definition also, and in units of solar radius, M, =DF/2, or DF = 2M\ ■ Finally,
AB == 1 + s — DF.
(8.3623-4)
In terms of known quantities, then
BC= 1,
AC = M2 =
DF = 2
L2
= 5,
AB= 1 +
Li + Li
L\ - m
Li + L2
U-L2 2(L] - m)
L\ +L2 L\ + L2
2m
U+L2
(8.3623-5)
466 EXPLANATORY SUPPLEMENT
The area of the solar disk covered by the Moon is given by
S = segment CFC' + segment CDC', (8.3623-g)
segment CFC' = 2(sector ACF - triangle ACE),
segment CDC' - 2(sector BCD - triangle BCE). (8.3623-7)
Using angles in radians:
Area of sector ACF = sA,
Area of sector BCD = B,
CE = 3 sin A = sin B,
AE - scosA,
EB = cos B. (8.3623-8)
Substituting into Equations 8.3623-7 and then 8.3623-6:
segment CFC' = 2 f i*2^ - ^s2 sin A cos A ),
segment CDC' - 2 ( -B — - sin B cos B
\2 2
S = (a2 A + B) — (s2 sin A cos A + sin B cos B)
= (s2A +B)-s sin C. (8.3623-9)
The area of the solar disk is 7r: hence the obscuration as a fraction of the disk is
S' = S/lT.
The angles A, B, and C may be evaluated from fundamental rules of trigonom-
etry. The working relations are:
cos C={L\+l^- 2m2) I (jLj - L\), 0 < C < n,
cosfi = (LiL2 +m2)lm(L\ + L2), 0 < B < tt, (8.3623-10)
A = n — (B + C).
If an observer wishes to construct a diagram showing the exact relationships of the
disks during the course of an eclipse, it is useful to have calculated these quantities
at a convenient tabular interval. Note that during central phase, all intersections
cease to exist and the quantities are undefined. During that phase, for an annular
eclipse S' = s2; for a total eclipse, S' = 1.
8 / ECLIPSES OF THE SUN AND MOON 467
g 363 Differential Corrections in Space and Time Differential corrections were in
times past used to adjust precisely calculated local circumstances to nearby loca-
tions. However, computing equipment has reached such a stage of advancement
that differentials have fallen into disuse in favor of repeating the whole calculation
for the new site.
Calculations are normally provided in provisional Universal Time (UT) or, more
precisely, UTl (see Chapter 2) using a predicted value of AT. For record purposes
this is not necessarily the best practice, but for the convenience of users it was
deemed significant enough to be necessary. If a later value of A T is adopted — such
that the offset becomes AT + 6T — then tabular quantities may be corrected quite
easily by applying
6X = - 1 . 002738 hT (longitude measured eastward) (8.363-1)
to all longitudes in the tabulated phenomena, and subtracting bT algebraically
from all tabulated times expressed in UT. Finally, interpolate the table back to the
original arguments.
For calculation of circumstances at elevations above the spheroid, the assumed
radius of the Earth is increased accordingly and the calculations repeated.
8.4 LUNAR ECLIPSES
8.41 Introduction
The calculation of lunar eclipses follows the same principles as that of solar eclipses.
The fundamental plane is perpendicular to the axis of the shadow, and the origin
of coordinates is the shadow's axis. The z-coordinate is not used. Since the observer
is on the body that is casting the shadow, the circumstances are the same for all
parts of the Earth from which the Moon is visible.
The criteria for the occurrence of a lunar eclipse, and the definitions of the
types that may occur, are given in Section 8.23.
8.42 Computations
Equations 8.2321-1 and 8.233-1 define specific values of the separation of centers
at which contacts occur (see Figure 8.2321.1). By conventional notation deriving
from solar eclipses, the angular distance between the centers of the Moon and the
shadow is designated L and at times of contact has values as follows:
at beginning and end of the penumbral eclipse, L] = f\ + sm\
at beginning and end of the umbral eclipse, L2 = fi + sm'<
at beginning and end of the total eclipse, £3 = fi - sm.
(8.42-1)
468 EXPLANATORY SUPPLEMENT
As is the case for solar eclipses, for dates when a lunar eclipse is to be calculated
apparent ephemerides of the Sun and the Moon are generated at a suitable interval,
by subtabulation if necessary. The usual interval is 10 minutes. The ephemeris is
corrected for the offset of the center of figure from the center of mass, as appropriate
for the chosen ephemeris, and the argument is changed to UT by applying AT.
8.421 Besselian Elements As in Section 8.322 and Equation 8.322-1, the geocen-
tric unit vector to the Sun, in an equatorial system, is
COS Q5 COS 6S
sin qs cos <5S
sin 6,
(8.421-1)
It follows that the vector of the antisolar point, along the axis, is rac = — rso. This
vector corresponds to g in Equation 8.322-2, with a = a% + 12", d = —Ss, for a
geocentric system and eliminates the step of solving for (a,d). To transform the
Moon's position to a geocentric system in which the z-axis is parallel to the axis
of the shadow, apply the same rotations as given in Equation 8.322-3, but for
units use seconds of arc. This calculation gives Equations 8.322-6, without the unit
conversion factor, and from which only (x,y) in the fundamental plane are used.
To summarize:
cos a cost/"
— cos as cos 6S
raG =
sin a cos d
= -rsG -
sin as cos 6S
(8.421-2)
sind
— sin Ss
«mF = Rj [~ - </) R3 (a + |) rmG =
cos 6m sin(am — a)
sin Sm cosd — cosi5m sindcos(am — a)
sin 6m sin d + cos 6m cos d cos(om - a)
«.[* y zf, (8.421-3)
where
dm --St,
a = as + n.
In the fundamental plane — i.e., where z = 0 — the separation
m = {x2+y2)1,
(8.421-4)
Call the vector form m. The hourly variation, m, is found by numerical differenti-
ation or, if a polynomial representation is used, by differentiating the polynomial.
g / ECLIPSES OF THE SUN AND MOON 469
The Besselian elements for a lunar eclipse are x, y, x, y, m, L\, Li, Z-3, f\ , f2. There
is no widely accepted convention, however, and the elements are not published in
The Astronomical Almanac.
g.422 Contact Times The contact times can now be found by either of two meth-
ods.
One method is to choose a time T0 near opposition, with corresponding m0, rh0.
For a contact at time T = Tq + r, in which / is positive or negative and L is chosen
from Equation 8.42-1,
(mo + mo t) ■ (m0 + iho t) = L • L. (8.422-1)
Equation 8.422-1 may be solved as a quadratic equation in t\ use the two roots
to estimate the two contacts. The two times may then be used to start a second
iteration, and convergence is very rapid.
Alternatively, the following discriminants may be tabulated as a function of
time:
Li — m,
Li — m,
Li - m, (8.422-2)
for a timespan starting before opposition such that initially m = 1.5Li, and contin-
uing until the first discriminant has passed through zero twice.
No approximations or manipulations are necessary. At each time step, the dis-
criminants are examined for a change of sign from the preceding step; when a change
occurs, find the time of the zero by inverse interpolation. The cases that may occur
appear in Table 8.422.1, and must occur in the sequence indicated.
In special cases where examination indicates that one phase might have begun
and ended within a single tabular interval, subtabulation or the alternative method
of polynomial iteration may be necessary.
8.423 Time of G reatest Obscuration The time of greatest obscuration occurs when
the axis of the shadow is closest to the center of the Moon, or when m is a minimum:
— (m • m) = 0 or m ■ in = 0, (8.423-1)
dt
The quantity
m ■ rh = xi: + yy (8.423-2)
may also be tabulated as a discriminant and the zero found by inverse interpolati
470
EXPLANATORY SUPPLEMENT
Table 8.422.1
Sequences and Conditions for Contact Times
Start
Condition
End
No eclipse
L\ - m < 0
No eclipse
Penumbral eclipse begins
lq — m =s o
Penumbral eclipse ends
Penumbral eclipse
i., - m > o and Lz - m < 0
Penumbral eclipse
Partial eclipse begins
L2 - m = 0 and L3 - m < 0
Partial eclipse ends
Partial eclipse
/_3 - m < 0
Partial eclipse
Total eclipse begins
£3 - m = 0
Total eclipse ends
Total eclipse
L3 - m > 0
Total eclipse
Note: The events in the leftmost column of the table occur in sequence from the top down, and in the
rightmost column of the table, from the bottom up.
8.424 Magnitude The magnitude is the fraction of the Moon's diameter covered
by shadow, in units of the lunar diameter. As published, magnitude corresponds
only to the maximum obscuration, but for use in special calculations, it may be
calculated as a function of time for both umbra and penumbra.
The expression is
Lj — m
—n , (= 1,2, (8.424-1)
in which L\ is used for penumbral magnitude and for umbral magnitude. Mag-
nitude may be greater than unity, but not negative.
8.425 Position Angles The position angle of contact P on the limb of the Moon is
the position angle M of m at the instant of contact, measured eastward (clockwise)
from the north; i.e.,
tanM = xl y
P = M for interior contacts
= Af + 1 80° for exterior contacts. (8.425-1)
8.426 Sublunar Points The latitudes (<p) and longitudes (A) of places that have
the Moon in the zenith at given times are given by
<P = 6m
A = Greenwich apparent sidereal time — am. (8.426-1)
8 / ECLIPSES OF THE SUN AND MOON 4J\
For a given position (^>o, An) found from this, the horizon circle is the locus of points
on a great circle that satisfy the condition:
tan <j> = - cot 4>0 cos(A0 - A). (8.426-2)
Values of A may be assumed and coordinate pairs calculated. The hemisphere de-
fined by this central point and horizon circle is where the Moon is above the horizon
at the instant.
8.5 TRANSITS
As mentioned in Section 8.1, a transit occurs when the shadow of one of the infe-
rior planets falls upon the Earth. Information concerning transits of Mercury and
Venus is tabulated in The Astronomical Almanac. This information consists of the
elements of the transit (positions and motions of the Sun and inferior planet at
time of conjunction), geocentric phases, and short formulas that are functions of an
observer's longitude and latitude to generate local circumstances. Additionally, a
world map (Mercator projection) is provided from which a user can determine the
region of visibility and approximate local circumstances.
Prior to and including the 1986 transit of Mercury, an extension of Newcomb's
(1882) heliocentric method was employed in the transit calculations for The As-
tronomical Almanac. The required heliocentric ephemeris of Mercury was based
on Newcomb's Tables of Mercury, and the heliocentric ephemeris of the Earth was
derived from Newcomb's Tables of the Sun. The corrections made to these ephemer-
ides and the actual method used to compute the transits is described in the original
Explanatory Supplement.
In theory, the general methods used to compute solar eclipse phenomena may
be extended to predict the circumstances of a transit once the inferior planet has
been substituted for the Moon. Chauvenet (1891) derived a special method, based
on improvements to Lagrange's method, which takes advantage of the small paral-
laxes of the Sun and the transiting planet. This method requires apparent geocentric
ephemerides of the Sun and planet as input. The apparent ephemerides can be read-
ily computed from a modern solar system ephemeris, such as DE200, using methods
described in Chapter 5. Chauvenet's method also allows for the generation of short
formulas, functions of longitude and latitude, from which local circumstances may
be computed. The interested reader may use the references given in Section 8.6 for
further details.
472
8.6 REFERENCES
EXPLANATORY SUPPLEMENT
Further information on calculations and predictions is often difficult to obtain. The fol-
lowing is a guide to the most important material.
Chauvenet, W. (1891). A Manual of Spherical and Practical Astronomy fifth ed. (J.B.
Lippincott Co., Philadelphia) reprinted 1960 (Dover Publications, New York) I, pp,
436-542 (solar eclipses); pp. 542-549 (lunar eclipses); pp. 591-601 (transits of inner
planets). This reference contains an account of the formulas necessary for the pre-
diction of the stated phenomena for the Earth generally, and for a particular place.
Various corrections derived from observations are also considered. The formulas and
constants are suitable for use with logarithms. An adaptation for modern use, upon
which much of the development in this chapter is based, is in the technical report by
Williams.
Her Majesty's Nautical Almanac Office (1961). Explanatory Supplement to the Astronom-
ical Ephemeris and the American Ephemeris and Nautical Almanac (H.M. Stationery
Office, London). Chapter 9 on eclipses and transits is the predecessor of the present
exposition.
Green, R. (1985). Spherical Astronomy (Cambridge University Press, Cambridge). This
book and the book by Murray are extremely helpful in presenting spherical trigonom-
etry in vector notation.
Herald, D. (1983). "Correcting Predictions of Solar Eclipse Contact Times for the Ef-
fects of Lunar Limb Irregularities" J. Brit. Astron. Assoc. 93, 241-246. This paper
gives precepts for constructing charts to correct solar eclipse contact times for the
effects of lunar limb irregularities. A useful discussion estimates the total effect and
its components.
Link, F. (1969). Eclipse Phenomena in Astronomy (Springer- Verlag, New York). This
book contains a wealth of information on lunar eclipses and transits; the informa-
tion is available almost nowhere else. The volume presents primarily physical and
atmospheric effects of the eclipsing body, but also includes computational and histor-
ical information. It also treats eclipse and occupation effects involving other planets,
natural and artificial satellites, radio wavelengths, and relativistic effects.
Liu Bao-Lin (1983). "Canon of Lunar Eclipses from 1000 B.C. to A.D. 3000" Pub. Purple
Mountain Obs. 2, 1. This volume contains data on 9800 lunar eclipses, including
penumbral. It contains the data in a different form from that of Meeus and Mucke,
and uses conventional augmentation of the Earth's shadow. The primary use of canons
is for chronological and historical research, but they are also useful for planning and
for seeking long-term cyclic features.
Meeus, J., Grosjean, C, and Vanderleen, W. (1966). Canon of Solar Eclipses (Perganion
Press, Oxford). In the style of Oppolzer, but with improved calculations, this canon
covers only solar eclipses from 1898 to 2510. An introduction contains the theory,
general description, and a discussion of eclipse cycles. Maps show the central tracks
over the entire Earth.
Meeus, J. and Mucke, H. (1983). Canon of Lunar Eclipses, -2002 to 2526 second ed. (As-
tronomical Office, Vienna). This book contains data on 10936 lunar eclipses, including
penumbral. The data are in the form of computer listings and plotted reference charts.
The data are in a different form from those of Liu, using Danjon's augmentation of
the Earth's radius. Correction for AT is not applied.
/ ECLIPSES OF THE SUN AND MOON
473
Mikhailov, A. (1931). "Uber die Berechnung der Breite der Totalitatszone bei Sonnenfin-
sternissen" Astronomisches Nachrichten 243, 51.
Mitchell, S.A. (1951). Eclipses of the Sun, fifth ed. (Columbia University Press, New York).
This book describes observing expeditions in the first half of the twentieth century.
Mucke, H. and Meeus, J. (1983). Canon of Solar Eclipses -2003 to +2526 (Astronomical
Office, Vienna). This book contains elements of 10774 solar eclipses, in the form of
computer listings and small plotted charts. The information is based on Newcomb's
solar theory and the ILE, and is arranged for use by Bessel's method. Correction for
AT is not applied.
Murray, C.A. (1983). Vectorial Astrometry (Adam Hilger Ltd., Bristol, England).
Newcomb, S. (1882). "Discussion and results of observations on transits of Mercury, from
1677 to 1881" APAEl, part VI (U.S. Naval Observatory, Washington).
Oppolzer, T.R.v. (1887). Canon der Finsternisse (Imperial Academy of Science, Vienna)
reprinted 1962 (Dover Publications, New York) with English translation. This book
contains elements of 8000 solar eclipses from -1207 November 10 (Julian proleptic
date) to 2161 November 17 (Gregorian date) and of 5200 lunar eclipses from -1206
April 21 to 2163 October 12. Maps show the central paths of total and annular eclipses
in the northern hemisphere. Data are arranged for calculation by Hansen's method,
using logarithms, and have known errors.
Smart, W.M. (1977). TextBook on Spherical Astronomy sixth ed. (Cambridge University
Press, Cambridge).
Williams, W. Jr. (1971). Prediction and Analysis of Solar Eclipse Circumstances National
Technical Information Service Technical Report NTIS No. AD726626 (Springfield,
Virginia).
CHAPTER
Astronomical Phenomena
by B.D. Yallop and C.Y. Hohenkerk
9.1 GENERAL ASPECTS OF THE NIGHT SKY
The configurations and phenomena of the Sun, Moon, and planets as seen from
the Earth arise from their apparent movement in the sky. Most of the phenomena
published in the "Diary of Phenomena" for example, are geocentric; the remainder
are heliocentric.
The Earth rotates about its polar axis once a day and produces an apparent
motion of the night sky about the celestial poles. In the Northern Hemisphere the
north celestial pole is elevated above the horizon. Facing away from the elevated
celestial pole, an observer sees the Sun, Moon, planets, and stars rise in the east and
set in the west. They reach their highest altitude as they cross the local meridian.
When the observer turns to face the elevated celestial pole, stars nearest the pole
neither rise nor set. They become circumpolar and cross the meridian each day
once above the pole at their highest altitude and once below the pole at their
lowest altitude. In the Northern Hemisphere, circumpolar stars appear to rotate
about the north celestial pole anticlockwise. In the Southern Hemisphere the effect
is reversed and they appear to rotate clockwise.
Superimposed on the diurnal rotation is an annual rotation caused by the
Earth's orbiting the Sun. Since the stars are seen by the naked eye after sunset, the
constellations appear to move from east to west, and to return to the same position
after a year. Relative to the Sun, the stars rise and set roughly four minutes earlier
each day. In the course of a month, the night sky appears to move two hours in
right ascension to the west. Also because of this orbital motion of the Earth, the
circumpolar stars in the Northern Hemisphere appear to rotate once a year in an
anticlockwise direction around the north celestial pole and in a clockwise direction
about the south celestial pole.
475
476 EXPLANATORY SUPPLEMENT
The Moon moves in an orbit inclined to the ecliptic by 6°; the Moon makes one
revolution about the sky from west to east in about a month. During this period
the phases of the Moon complete a cycle from new to full and back to new. The
orbit of the Moon is moving around the ecliptic, so that other aspects of the Moon's
position in the sky, such as its maximum and minimum declination, change from
one month to the next.
It is important to know when the planets are in the most favorable position for
observation. The outer planets, for example, are best seen around opposition. They
are in their most unfavorable position around conjunction. The inner planets are
different — they are in their most favorable position near greatest elongation, even
though they are not at full phase. At superior conjunction the phase is around full,
but the planets are difficult to see because they are furthest from the Earth and
usually too close to the Sun. At inferior conjunction the inner planets are nearest
to the Earth, but again they are difficult to see because their phase is small, and
they are too close to the Sun.
Often the times of phenomena need not have any great precision; sometimes the
nearest hour or even the nearest day are sufficient for observational purposes. The
dates and times, however, usually depend on the coordinate system. For historical
reasons the conjunctions and oppositions of planets have always been calculated in
geocentric ecliptic coordinates. On the other hand, the conjunctions of planets with
other planets, bright stars, or the Moon have always been calculated using equato-
rial coordinates; the phenomena are then observed more easily with an equatorially
mounted telescope. In some cases the times of phenomena have been defined as the
maxima or minima of the distances from the Sun or the Earth or the elongation
from another body. In such cases, the phenomena are independent of the coordinate
system.
9.2 CONFIGURATIONS OF THE SUN, MOON, AND PLANETS
The Universal Times (UT) of the principal astronomical phenomena involving con-
figurations of the Sun, Moon, and planets are given in The Astronomical Almanac
pages Al-All under the heading "Phenomena". In most cases the times are given
to the nearest hour, but for certain heliocentric phenomena of the planets, only
the date appears. The times of the phenomena of greatest general interest — the
beginning of the seasons and the phases of the Moon — are to the nearest minute,
although the accuracy is of no significance.
Times may be calculated using a daily tabular ephemeris. For quantities that
vary slowly with time, such as planetary apsides, a tabular interval of five days gives
better results. On the other hand, quantities such as the phases of the Moon vary
rapidly, and a shorter interval of a half or quarter day is more appropriate. If the
ephemeris is in TDT, then, to convert to UT, add the correction AT" = UT - TDT
to the time obtained at the end of the calculation.
9 / ASTRONOMICAL PHENOMENA 477
The time t of the phenomenon derives from a functional expression involving
the parameters associated with the phenomenon. The time of the phenomenon is
calculated from the equation /(f) = 0 or f'(t) = 0, where f(f) = df(t) I dt.
9.21 Interesting Phenomena of the Sun, Earth, and Moon
9.211 Equinoxes and Solstices The times of the equinoxes and solstices are de-
fined to be when the Sun's apparent ecliptic longitude As is a multiple of 90°; i.e.,
it is calculated from /(f) - 0, where /(f) = Xs - 0°. 90°, 180°, or 270°. Thus in
the northern hemisphere, for the spring equinox /(f) - Xs, for the summer solstice
f(t) = Xs — 90°, for the autumn equinox /(f) = Xs - 180° and for the winter solstice
f(t) = Xs — 270° . At the equinoxes the Sun crosses the equator when the length of the
day exceeds the length of the night due to refraction, semidiameter, and parallax
of the Sun. At that time the lengths of the day and night are approximately equal
everywhere.
The time of the commencement of the seasons shows a progressive change
because of the leap-year cycle. Because the period of revolution of the Earth about
the Sun is not commensurate with the Gregorian calendar year, it is only after a
complete cycle of four centuries that the seasons again commence at approximately
the same times. In the present century the latest dates for the seasons occurred in
1903, and the earliest will be in 2000; by the year 2096 the seasons will begin at
their earliest possible times {see Table 9.211.1).
Table 9.211.1
Time of Commencement of the Seasons
Spring
Summer
Autumn
Winter
d h
d h
d h
d h
Latest
1903
March 21 19
June 22 15
Sept. 24 06
Dec. 23 00
2000
20 08
21 02
22 17
21 14
Earliest
2096
19 14
20 07
21 23
20 21
Note: The total range in times is about 54 hours in each case.
9.212 Perihelion and Aphelion of the Earth In the "Diary of Phenomena,1' the
times when the Earth is at perihelion (the Sun is at perigee) are defined to be
those for which the Sun's geometric distance R is a minimum. Likewise, the times
when the Earth is at aphelion (the Sun is at apogee) are defined to be when R is
a maximum. Thus if t is the time of perihelion, it is calculated from the equation
f'{t) = 0 where /(f) = R. The times do not always agree with those calculated from
the times when the Sun is at perigee or apogee in its mean elliptical orbit (i.e., the
478 EXPLANATORY SUPPLEMENT
time t when the mean longitude of perigee F = 0, or apogee F — 180° = 0) because
of perturbations by the planets.
9.213 Phases of the Moon The times of the phases of the Moon are tabulated to
the nearest minute of UT in The Astronomical Almanac pages Al and Dl and are
given to the nearest hour of UT in the "Diary" on pages A9-A11. They are the
times when the excess of the Moon's apparent geocentric ecliptic longitude XM over
the Sun's apparent geocentric ecliptic longitude is 0°, 90°, 180°, or 270° — i.e., the
times when /(f) = 0 where /(f) = XM - Xs for new Moon, /(f) = XM - Xs - 90° for
first quarter, /(f) = \M - Xs - 180° for full Moon and /(f) = XM - Xs - 270° for last
quarter. Because the times are determined from geocentric coordinates, they are
independent of location on the Earth.
Owing to the rapid variations in the distance and velocity of the Moon, the
intervals between successive phases are not constant, and it is not possible to check
these times by differencing. Moreover, there is no simple prediction formula with
which to make a comparison. Examination of the higher differences of the successive
times of the same phenomenon provides a check.
The phases of the Moon do not recur on exactly the same dates in any regular
cycle, but the approximate dates of the phases in any year can be found from
the dates on which the phase occurred 19 years previously. Thus in the Metonic
cycle, in which 19 tropical years are nearly equal to 235 synodic months (new moon
to new moon), the phases recur on dates that are the same or differ by one or
occasionally two days, depending on the number of intervening leap years and on
the perturbations of the Moon. For example, during 2000 the dates are the same
as in 1981 on thirty occasions, and differ by one day for the remaining nineteen.
Another relevant cycle is the Saros, which consists of 233 synodic months and
equals nearly 19 passages of the Sun through the node of the Moon's orbit; not
only will the moon phases recur but eclipses as well. Moreover in this cycle the
Moon's apse makes 239 revolutions and returns to nearly the same position, so the
durations of solar eclipses are similar as well.
9.22 Geocentric Phenomena
The times of geocentric phenomena are calculated from the expressions for /(f)
listed below. Table 9.22.1 contains the phenomena for which the time is obtained
from the equation /(f) = 0. Table 9.22.2 contains phenomena for which the time is
obtained from the equation /'(f) = 0. In this table, except for the stationary points
in right ascension, the phenomena are independent of the coordinate system.
The notation A, a, 6 refers to the apparent geocentric longitude, right ascension,
and declination; and the subscripts S, M, p. b refer to the Sun, Moon, planet, or
bright star, respectively. The distances r, R, A, and p refer to the geometric distances
of the Sun-planet, Earth-Sun, Earth-planet, and Earth-Moon, respectively; 7r is
/ ASTRONOMICAL PHENOMENA
Table 9.22.1
Geocentric Phenomena for which /(f) = 0
Phenomenon
/(f)
Remarks
Seasons:
Vernal Equinox
Summer Solstice
Autumnal Equinox
Winter Solstice
As
As - 90°
As- 180°
As - 270°
Conjunctions ol the planets with the Sun:
Inferior planets:
interior conjunction
superior conjunction
Superior planets:
conjunction
opposition
V
As
As
As
180°
Phases of the Moon:
New Moon
First Quarter
Full Moon
Last Quarter
Ap - As
Am - As
A„ - AS - 90°
A„-AS^180°
AM - AS - 270°
Conjunctions of other bodies:
planet with planet »p, - "p?
planet with bright star ap - ab
Moon with planet "u - ap
Moon with bright star otM - ab
March equinox, First Point of Aries
June solstice
September equinox
December solstice
Ap - As changes from plus to minus
Ap - As changes from minus to plus
separation = 6Pl - fipj
separation = 6P — 6/,
separation = % — Sp
separation = 6M - Sb
Table 9.22.2
Geocentric Phenomena for which / (f) = 0
Body
Phenomenon
fit)
Remarks
Mercury Greatest elongation
Venus
Venus
Earth
Moon
Mars
Planet
Greatest brilliancy
Apsides
Apsides
Closest approach
Stationary in
right ascension
-28° < E < +28°
-47° < E < +47°
(r + A + R)(r+A- R) Q gg AU < A < o. 47 AU
r=R
p = 1 / sin w
A
At perihelion r is a minimum
At aphelion r is a maximum
At perigee p is a minimum
At apogee p is a maximum
At perigee jt is a maximum
At apogee * is a minimum
480
EXPLANATORY SUPPLEMENT
the horizontal parallax of the Moon. The five bright stars used in the "Diary" are
Atdebaran, Pollux, Regulus, Spica, and Antares.
The following formulas are useful for calculating elongation and longitude:
cos E = sin Sp sin <SS + cos 6p cos <5S cos(ap — as), (9.22-1)
sin e sin 6 + cos e cos 6 sin a
tan A = , (9.22-2)
cos o cos a
where e is the true obliquity of the ecliptic.
9.221 Visibility of Planets Table 9.221.1 shows the criteria adopted for the "Di-
ary" for the minimum elongations from the Sun at which Mercury, the bodies used
in navigation, and the minor planets can be seen with the naked eye.
Table 9.221.1
Visibility Criteria for Geocentric Phenomena
Body
Minimum Elongation
Moon, Mars, Saturn
15°
Minor planets
15°
Venus, Jupiter
10°
Mercury
10° + visual magnitude of Mercury
9.222 Synodic Periods Once the date and time of a geocentric phenomenon are
known, the approximate time of the next similar phenomenon may be estimated by
adding the synodic period T of the body concerned. The eccentricities of the orbits
of the Earth and the body, and the perturbations by other bodies, will introduce
errors in this estimate. Table 9.222.1 gives synodic periods of the major planets and
Table 9.222.1
Synodic Periods of the Planets and First Asteroids
Planet
Days
Planet
Days
Asteroid
Days
Mercury
116
Saturn
378
Ceres
467
Venus
584
Uranus
370
Pallas
466
Mars
780
Neptune
367
Juno
474
Jupiter
399
Pluto
367
Vesta
504
9 / ASTRONOMICAL PHENOMENA
481
Table 9.222.2
Long-Period Cycles of Mercury, Venus, and Mars
For Mercury
54 sidereal periods =
= 13 years +
2 days
137 sidereal periods =
= 33 years —
1 day
For Venus
13 sidereal periods
: 8 years —
1 day
359 sidereal periods =
= 243 years -
I day
For Mars
8 sidereal periods
15 years +
1 7 days
17 sidereal periods
32 years -
9 days
25 sidereal periods =
47 years +
8 days
42 sidereal periods =
79 years -
1 day
the four brightest minor planets. The periods may be used for a systematic search
for geocentric phenomena.
The synodic period T is calculated from the mean motion of the planet, np, and
the mean motion of the Earth, ne. Thus,
360^ 360^
T ~ — for inner planets, and T for outer planets. (9.222-1)
np — ne ne — np
For the slowly moving planets (Jupiter to Pluto) and the minor planets, the
error in using these values is small, but for Mercury, Venus, and Mars, the mean
synodic period is less useful. Much more accurate estimates of when these planets
may be seen again in the same part of the sky may be made by using long-period cy-
cles that contain, with varying degrees of accuracy, integral numbers of revolutions
of the Earth and the planet. (See Table 9.222.2.)
Any particular phenomenon of a planet repeats itself after each cycle at the
same time of year and in the same part of the sky; However, repetition does not
occur after a single synodic period. For Venus a useful and accurate form of the
relation is
5 mean synodic periods = 8 calendar years — 214, (9.222-2)
unless the interval includes a century year that is not a leap year.
9.23 Heliocentric Phenomena
Certain heliocentric phenomena of the planets are given in The Astronomical Al-
manac, page A3. The dates of perihelion and aphelion are those on which the
geometric distance of the planet from the Sun is a minimum and maximum re-
spectively; the dates are thus the times when the first derivative of the distance is
zero — j _e-5 f'(t) - drldt = 0. Owing to the presence of perturbations in the planetary
482
EXPLANATORY SUPPLEMENT
Table 9.23.1
Heliocentric Phenomena
Phenomenon /(f) Type Remarks
Nodes b /(f) = 0 At the descending node, b is decreasing.
At the ascending node, b is increasing.
Apsides r /'(f) = 0 At aphelion, r is a maximum.
At perihelion, r is a minimum.
Greatest latitudes b /'(f) = 0 At greatest latitude North, b is positive.
At greatest latitude South, b is negative.
motion, these dates may differ from those obtained from the angular elements of the
mean orbits. The actual disturbed motion of the planets is also used to determine
the dates when they pass through the nodes on the ecliptic, and when they reach
greatest latitudes north or south. At the nodes, the heliocentric latitude is zero; i.e.,
/(f) = b = 0. The value of /(f) changes from negative to positive at the ascending
node, and from positive to negative at the descending node. These dates are given
each year for Mercury, Venus, and Mars, but they occur less frequently for the
other planets, and in these cases are given as additional notes when necessary. The
dates on which a planet has its greatest latitude north or south is determined as
the times at which the first derivative of the latitude is zero— i.e., /'(/) - dbldt = 0.
Table 9.23.1 shows /(f) for the various phenomena.
9.3 RISINGS, SETTINGS, AND TWILIGHT
The astronomer is concerned with the phenomena of rising, setting, and twilight
primarily in regard to the planning of observations. Precision of better than a minute
of time is not required for this purpose, and extensive tables of these phenomena
are in The Astronomical Almanac, pages A14-A77.
The tabulated times of the phenomena refer to sea level with a clear horizon and
an adopted correction for refraction under normal meteorological conditions. The
actual times of rising and setting may differ considerably, especially near extreme
conditions when the altitude is changing slowly. This difference can reach half a
minute of time at mid-latitudes and more at high latitudes. The illumination at
the beginning or end of twilight also varies greatly with meteorological conditions.
Precise times have little real significance, except in special circumstances, such as
navigation at sea.
No data are given for the times of rising and setting of planets. Data may be
obtained fairly simply from navigation tables such as the Sight Reduction Tables
9 / ASTRONOMICAL PHENOMENA 483
for Air Navigation. Within their range of declination (0°-29°), these tables may
also be used to find the times of rising and setting of stars.
Section 9.33 gives algorithms that may easily be programmed into a small
calculator or computer to calculate the times of the phenomena.
9.31 Sunrise, Sunset, and Twilight
The data given in The Astronomical Almanac enable the times of sunrise, sunset,
and the beginning and end of civil, nautical, and astronomical twilight to be found
for any position between latitudes 66° north and 55° south. The times, tabulated
for every fourth day, are the local mean times of the phenomena on the meridian
of Greenwich and in the specified latitude; interpolation is necessary to obtain the
local mean times for intermediate latitudes, for intermediate days, and for longitude.
To an accuracy of about five minutes this interpolation can generally be done by
sight; near limiting conditions, when interpolation becomes difficult, large changes
of time correspond to only small changes in depression and accurate times have
little real meaning.
Interpolation for latitude is nonlinear. Interpolation for longitude, which is
rarely justified, can be combined with the interpolation for date by increasing for
west longitudes, or decreasing for east longitudes, the Greenwich date by the ap-
propriate fraction (longitude in degrees/360); for sunrise and sunset, the error due
to neglecting the variation with longitude amounts to a maximum of two minutes
in latitudes up to 60°. The times so obtained are local mean times, which can
be converted to universal time by applying longitude in time — adding if west and
subtracting if east. Standard times are obtained by adding (subtracting) to the
local time four minutes for every degree of longitude west (east) of the standard
meridian.
At the tabulated times of sunrise and sunset, the geocentric zenith distance
of the center of the Sun is 90°50'; i.e., a geocentric altitude of -50', of which 34'
is allowed for horizontal refraction and 16' for semidiameter; the Sun's apparent
upper limb is thus on the horizon. Corrections are necessary if some other value of
the altitude is required, (such as for the conventional meteorological value of -34 )
or to allow for the height of the observer and the elevation of the actual horizon.
In such cases use Equation 9.31-1.
At the times given for civil, nautical, and astronomical twilight, the altitude of
the center of the Sun is -6°, -12° and -18° respectively. These tabulations will
produce times for any desired altitude between -50' and —18°.
The Nautical Almanac and The Air Almanac tabulate the times of sunrise,
sunset, and the beginning and end of civil twilight (altitude -6°) every three days
for latitudes between 72° north and 60° south; times of the beginning and end
of nautical twilight (altitude -12°) are also in The Nautical Almanac. The Air
Almanac provides graphs that enable corrections to the time of sunrise or sunset.
484 EXPLANATORY SUPPLEMENT
These corrections give the times at which the Sun has altitudes between 0° and
— 12° (that is, depressions to 12°).
Sunrise or sunset at a height H meters above the level of the horizon occurs
when the Sun's altitude is approximately
-50' -2'.12V7/ (1'.17 fori* in feet), (9.31-1)
so that the same table gives corrections for height to the times of sunrise and sunset.
In The Air Almanac the corrections can be obtained from graphs. The constant 2. 12
is really a slowly decreasing function of H, but below a height of about 5000 m it is
constant; it decreases to a value of 2.00 at 100000m.
Times of rising and setting and associated phenomena change rapidly from
day to day in polar regions or may not occur for long periods, the Sun being
continuously above or below the horizon; accurate times are therefore difficult to
tabulate. Diagrams are given in The Air Almanac that enable approximate times
that are sufficiently accurate for all practical purposes to be obtained.
9.311 A Low Precision Ephemeris for the Sun The following algorithm gives the
equation of time (E), the Greenwich hour angle (GHA), declination (6), and semidi-
ameter {SD) of the Sun, in degrees, to a precision of better than l'.O, which can
be used with the algorithm for rising and setting (see Section 9.33). The equation
of time is also useful for erecting sundials and determining the transit time of the
Sun.
(1) Using the Jul ian date, JD, and the universal time, UT, in hours, calculate
T, the number of centuries from J2000:
T = (JD + UT / 24 - 245 1545. 0) / 36525. (9.3 1 1-1 )
(2) Calculate the Solar arguments; the mean longitude corrected for aberration,
L; the mean anomaly, G; the ecliptic longitude, A; and the obliquity of the
ecliptic, e:
L = 280f 460 + 36000f 770 T, remove multiples of 360°,
G = 357f 528 + 35999f 050 T,
A = L+ lf915sinG + Of 020 sin 2G,
e = 23?4393 -0?01300r. (9.311-2)
9 / ASTRONOMICAL PHENOMENA
485
(3) The equation of time E, the GHA, 6, and SD are given by
E = - If 915 sin G - Of 020 sin 2G + 2f 466 sin 2A - Of 053 sin 4A,
GHA = 15UT - 1 80° + £,
b = sin- '(sine sin A),
SD = Of 267/(1 — 0.017 cos G). (9.311-3)
Plotting the position of the apparent Sun relative to the mean Sun (which moves
Uniformly along the equator) produces the analernmic curve, see Figure 9.31 1.1. The
displacement of the apparent Sun in longitude is given by the equation of time (E)
and the displacement in latitude is given by the declination (6). The progress of
the apparent Sun throughout the year is indicated at the beginning of each month
on the closed curve. The analernmic curve plays an important role in certain types
of sundial.
9-32 Moonrise and Moonset
The tables in The Astronomical Almanac give, for every day and for a range of lat-
itudes from 66° north to 55° south, the local mean times of moonrise and moonset
486 EXPLANATORY SUPPLEMENT
for the meridian of Greenwich. Interpolation for latitude and longitude is necessary
to obtain local times for other places. In practice, times are rarely required more
accurately than to within about five minutes, and interpolation can be done men-
tally. Formal interpolation, using tables such as are given in The Nautical Almanac
and The Air Almanac, yields times accurate to about two minutes. The times ao
obtained may be converted to universal time or standard time by applying longi-
tude in time — adding if west and subtracting if east. Times for latitudes between
72° north and 60° south are given in The Nautical Almanac and The Air Almanac.
In The Air Almanac times appear in graphical form above 72° north, right up to
the North Pole.
In calculating the times of moonrise and moonset, the true altitude h of the
center of the Moon is
h = —34' — semidiameter + horizontal parallax (9.32-1)
where 34' is allowed for horizontal refraction; depending upon the distance of the
Moon, h lies in the range 5' to 11'. At these times, the upper limb of the Moon is
on the horizon; no allowance is made for phase. At a height of H meters above the
horizon, the altitude of the Moon when its upper limb is on the horizon is decreased
by 2'A2Vh (see Section 9.331).
The Moon revolves round the Earth and makes one complete revolution relative
to the Sun in a synodic month of mean length 29.53 days; in that time the Moon
therefore appears to lose one transit across any meridian and, in general, one rising
and one setting. During each month there is therefore no moonrise on one local day
(near last quarter) and no moonset on one local day (near first quarter). In high
latitudes the times of the phenomena change rapidly from day to day and may not
occur for long periods, the Moon being continuously above or below the horizon;
in these extreme conditions the times of moonrise and moonset sometimes decrease
from day to day, instead of the usual increase in lower latitudes, and it is possible
to have two moonrises or two moonsets during the same local day.
9.33 Formulas Associated with Rising and Setting
Consider first an algorithm for calculating the times of rising or setting of a body
at true altitude h.
The GHA and declination (6) of the body are required to about one minute of
arc as a function of UT. For the Moon, the horizontal parallax 7r is also required.
To speed up the calculation, which is iterative, it is convenient to express GHA, o,
and 7r as daily polynomials. For the Moon second-order polynomials are required for
GHA and 6, and first order for tt. For other bodies like the Sun and the inner planets,
first-order polynomials are sufficient. For stars and outer planets, a constant daily
value for declination may be adopted. Alternatively low-precision formulas for the
9 / ASTRONOMICAL PHENOMENA 487
ephemeris of the Sun (Section 9.311) may be used. For the Moon, the low-precision
formulas on page D46 of The Astronomical Almanac may be used.
At sunrise and sunset the apparent altitude of the upper limb on the horizon
is zero and hence the adopted true altitude in degrees is h = -50 / 60 - 0. 0353 i/r7
where H is the height, in meters, of the observer above the horizon.
Similarly at moonrise and moonset the apparent altitude of the upper limb on
the horizon is zero and the adopted true altitude h = -34/60+0. 7275?r -0.0353^.
At rise and set of a star or a planet, the apparent altitude of the body is zero; hence,
h- -34 / 60-0. 0353
Other important cases to consider are twilights: civil twilight h = -6°, nautical
twilight h - -12°, and astronomical twilight h = -18°.
The time of rise or set, UT, in hours, is found by solving iteratively the equation
UT = UT0 - (GHA + A ± /)/ 15, (9.33-1)
where the plus sign is used for rise and the minus sign for set, and t is the hour
angle of the body at UTo, which is given by
sin/i - sin <p sin 6
cosf= ; » — (9.33-2)
cos (p cos o
and <p is the latitude and A the longitude (positive to the east).
If cos t > +l,set t-0°;
(9.33-3)
If cosf < -l,set t= 180°.
As an initial guess, set UT0 = 12", although any value in the range 0" to 24" will
do. After each iteration, add multiples of 24h to set UT in the range 0h to 24h.
Replace UT0 by UT until the difference between them is less than 0?008. If for
several iterations cosf > 1, it is likely that there is no phenomenon and the body
remains above the true altitude h all day. On the other hand, if cos / < - 1 , the body
remains below the true altitude all day. In each lunation, around first quarter, there
is always a day when the Moon does not set, and another, around last quarter, when
the Moon does not rise.
At latitudes above 60° the algorithm may fail, and it is necessary to use a more
systematic approach. If hQ is the adopted true altitude, the times when h = ho where
sink - &in4> sin 6 + cos 4> cos 6 cos(GHA + A) (9.33-4)
are the times of rising or setting.
These roots, if they exist, will lie between the true altitude at 0h and 24" and
the maximum and minimum altitudes during the day. The maximum and minimum
488 EXPLANATORY SUPPLEMENT
altitudes occur at or very near upper and lower transit. These times are found by
setting t = 0 and t = 180 in Equation 9.33-1 and iterating.
Only for the Moon at high latitudes is it necessary to calculate the maximum
and minimum altitudes more precisely. This calculation is done by fitting a second-
order polynomial to the true altitude at points around upper or lower transit and
differentiating with respect to time to find the turning point.
9.331 The Effects of Dip and Refraction Dip is required for reducing sights made
with a marine sextant and for calculating the true altitude of a body at rise and
set. Dip is the apparent angle between the geometric horizon (at right angles to the
vertical) and the visible horizon. For an observer at height H above the surface of
the Earth, dip D(H) = \'.15\/H, where H is in meters (0'.97\/7/ if H is in feet) (see
Woolard and Clemence, 1966, page 215).
A navigator using a marine sextant sets on the apparent horizon and wants
to know how much the horizon appears to be below the horizontal and therefore
only needs to know the dip. On the other hand, when computing the times of rise
or set of a body, we need to know the true altitude when the body appears to be
on the horizon, which involves dip and the refraction between the horizon and the
observer R(H), where R(H) - 0'.31\/H and H is in meters (or 0'.20\/77 if H is in feet).
In Figure 9.331.1, O is the observer at height H above the surface of the Earth.
S is the body at rise or set, and A is the observer's horizon. The true altitude of S
from the vantage point A is —34' due to refraction. The true altitude of 5 from O is
the true altitude at A - D(H) - R(H) = -34' - V.15y/H - Q'.31Vh = -34' - 2'A2y/H.
Figure 9.331.1
The horizon at rising or setting
g / ASTRONOMICAL PHENOMENA
489
Refraction, the bending of light as it passes through the atmosphere, is illus-
trated in Figure 9.331.1 for the rising or setting of a celestial object. The follow-
ing approximate formulas (Bennett, 1982), are valid for all altitudes and standard
meteorological conditions. More precise algorithms are given in Section 3.28. The
amount of refraction, R, is given by
R(hu) =
0?0167
tan(Aa + 7.31/(fta+4.4))'
where h„ is the apparent altitude. Alternatively, if ft is the true altitude, then
0?0167
(9.331-1)
R(h) =
tan(/i + 8.6/(/i + 4.4))"
(9.331-2)
9.332 Time for a Specified Shadow Length The lengths of shadows depend upon
the apparent altitude of the Sun, ha. The true altitude h of the Sun is needed to
find the required time where
h = ha-&{K) (9.332-1)
and R is the refraction at apparent altitude ha which may be calculated approxi-
mately using Equation 9.331-1. Alternatively, if h is known and ha is required use
Equation 9.331-2 for calculating the refraction at true altitude h and set
ha = h+R(h).
(9.332-2)
9,333 Rates of Change The rate of change of hour angle with altitude may be
obtained by differentiating the altitude as shown in Equation 9.33-4. Assuming
that A, 4>, and 6 do not change with time, then the change in time, (At), in hours,
for a given change in altitude (Ah), in degrees, can be calculated from the simple
expression
1
At =
A LHA coso sinZ
Ah.
(9.333-1)
where A LHA is the rate of change of hour angle per hour and Z is the azimuth.
For the Sun, A LHA = 15° per hour and for the Moon, A LHA = 14?493 per hour.
Near rise and set when h ~ 0°, this expression can be further simplified to
At = A Ah,
where A = 60 / ( A LHA v^cos2^ - sin2 6 ) ,
(9.333-2)
and At is now in minutes of time. This correction is accurate to within a few minutes
provided A does not exceed 20 and Ah does not exceed a few degrees.
490 EXPLANATORY SUPPLEMENT
9.334 Time for a Specified Azimuth Another interesting problem is to calculate
the time when the azimuth Z of a body takes a particular value. The time UT is
calculated iteratively from
UT = UT0-(GHA + A±r)/ 15, (9.334-1)
where the plus sign is used if Z < 180° and the minus sign is used otherwise, t is
calculated from
-ab±y/\ + a2-b2
cos t = - = , (9.334-2)
1 + a1
where a = tan Z sin <j>, b = - cos <f> tan 6 tan Z, and the plus sign is used if 90° < Z <
270°; otherwise use the minus sign. In the special cases Z - 90° and Z = 270° or if
1 + a2 - b2 < 0, then
cos t = — b I a = tan 8 1 tan <t>;
If cosf>+l, set cosr = +l; (9.334-3)
If cos t < — I, set cos r - — I.
Iterate Equation 9.334-1 until UT differs from UT0 by less than 0f008.
9.335 Times of Transit Times of transit are required to low precision for planning
observations, and for setting instruments. The times for a particular place may be
calculated for any body by iterating Equation 9.33-1 and setting / = 0°. For the
Sun, the transit time may be determined from 12h — El 15, where E is the equation
of time given in Section 9.311.
9.34 Illumination
The ground illumination (i.e., the illuminance on a horizontal surface) from nat-
ural sources varies considerably during the day. The variation of the illumination
on a horizontal surface, in clear conditions, on the surface of the Earth is shown
diagrammatically in Figure 9.34.1 as a function of altitude.
This section contains a simple method, suitable for use with electronic cal-
culators, of calculating ground illuminance from sunlight, twilight, and moonlight
based on data published by the Radio Corporation of America (RCA Electro -Optics
Handbook, 1974). A more precise method is available in United States Naval Obser-
vatory CircularNo. 171. This method is based on the extensive Natural Illumination
Charts produced by Brown.
Data from the handbook has been represented by sets of cubic polynomials
for the Sun and the full Moon at mean distance, for various ranges of altitudes,
9 / ASTRONOMICAL PHENOMENA
491
-20 0 +20 +40 +-60 +80
Altitude
Figure 9.34.1
Ground illumination from various sources
in Table 9.34.1. The illuminance is measured in lux or lumens per square meter;
however, the logarithm of the illuminance has been used because large ranges are
involved, and also the sensitivity of the human eye varies in proportion to the
logarithm of intensity. Thus the illuminance / in lux is given by
log,0/ = k + k* + hx2 + hx3, (9.34-1)
where x = h I 90, h is the altitude, and /0, l{, l2, 1} are the appropriate set of
coefficients given in Table 9.34.1 for various ranges of altitude. To obtain the total
492
EXPLANATORY SUPPLEMENT
Table 9.34.1
Coefficients for Calculating Ground Illumination
From the Sun
Maximum
Altitude range
to
•1
i
•2
fa
error
20°
90°
3.74
3.97
-4.07
1.47
0.02
5
20
3.05
13.28
-45.98
64.33
0.02
-0.8
5
2.88
22.26
-207.64
1034.30
0.02
-5
-0.8
2.88
21,81
-258.11
-658.36
0.02
-12
-5
2.70
12.17
-431.69
-1899.83
0.01
-18
-12
13.84
262.72
1447.42
2797.93
0.01
From the Full Moon at Mean Distance
20°
90°
-1.95
4.06
-4.24
1,56
0.02
5
20
2.58
12.58
-42.58
59.06
0.03
-0.8
5
-2.79
24.27
-252.95
1321.29
0.03
ground illumination from the Moon, corrections for phase and parallax must also
be added (see Section 9.342).
The values for the Sun in daylight give the illuminance from direct and indirect
sunlight in a cloudless sky. The values for indirect sunlight are about a factor of
10 smaller. If the sky is overcast, the values for indirect sunlight should be reduced
by about a factor of 10, and if the sky is very dark by about a factor of 100. A
correction of 21og|0(5/ 50), where S is the semidiameter of the Sun and S0 is the
semidiameter at mean distance, should be added to the daylight illuminance to
allow for the variable distance of the Sun from the Earth.
When the Sun is below the horizon, the sky is illuminated from the sources
discussed in the following sections.
9.341 Twilight This is caused by the scattering of sunlight from the upper layers
of the Earth's atmosphere. It begins at sunset (ends at sunrise) and is conven-
tionally taken to end (or begin) when the center of the Sun reaches an altitude
of -18°. At an altitude of -18°, astronomical twilight, the indirect illumination
from the Sun on a horizontal surface is about 6 x 10~4 lux, rather less than the
contribution from starlight and airglow. The actual brightness of the sky depends
on where the observer looks and where the illumination is coming from, as well as
on meteorological conditions.
In navigational practice and for certain civil purposes, two intermediate steps
in the twilight period are recognized and tabulated: civil twilight ends (or begins)
when the Sun reaches an altitude of -6° and nautical twilight ends (or begins)
when the Sun reaches an altitude of -12°. Before morning and after evening civil
twilight, outdoor activities that depend on natural lighting require artificial illu-
mination. The degree of illumination at the beginning and end of civil twilight (in
9 / ASTRONOMICAL PHENOMENA 493
good conditions and in the absence of other illumination) is usually described for
navigational purposes as illumination such that the brightest stars are visible and
the sea horizon is clearly defined; for the beginning and end of nautical twilight,
the corresponding statement is that the sea horizon is in general not visible and it
is too dark for the observation of altitudes with reference to the horizon.
9.342 Moonlight The illumination received from the Moon varies according
to phase, altitude, and atmospheric extinction. From full Moon in the zenith, the
intensity of illumination on a horizontal surface is approximately 0. 27 lux, equiv-
alent to that from the Sun at an altitude of about —8°. The amount of ground
illumination at night from moonlight — log10A/, where M is the illuminance in lux —
may be calculated from L\, the ground illumination from a Full Moon at altitude
h, at a mean distance that has been corrected for phase (Li) and parallax (L3) as
follows:
log!0 M = L, + L2 + £3. (9.342-1)
Extracting the appropriate set of /'s from the table and setting x = h I 90, where h
is the altitude of the Moon, then
Li - k + l\x + I2X2 + /3JC3,
L2 = -8.68 x 10_3/-2.2x \0~9f, (9.342-2)
L3 = 21og10(7r/0.951),
where /=180-E
and cos E = sin 6<, sin &m + cos 6s cos 6m cos(ctM — as)-
Figure 9.34.1 also shows the ground illumination on a horizontal surface from the
standard phases of the Moon at mean distance and at various altitudes. The half
Moon gives only one-ninth as much light as the full Moon. For astronomical obser-
vations, the position of the Moon in the sky, rather than the general illumination,
is the most important factor.
9.343 Starlight, Airglow, Aurora, Zodiacal Light, and Gegenschein The total illu-
mination from the stars contributes about 2. 2 x 10 4 lux, rather less than the Sun
at the beginning and end of astronomical twilight. The illumination from the stars
together with airglow contributes about 2 x 10~3 lux. The illumination from the
aurora (which depends upon solar activity) may in rare cases be comparable with
moonlight. The other sources are very faint and never give an illumination greatly
exceeding that from starlight. As with moonlight, the position of the source is the
most important factor for astronomical observations.
494
9.4 OCCULTATIONS
EXPLANATORY SUPPLEMENT
Only dates and general areas of visibility for the occultations of planets and bright
stars by the Moon are published in The Astronomical Almanac. Ranges of dates of
occultations of X-ray sources by the Moon are also given. The International Lunar
Occupation Centre, Astronomical Division, Hydrographic Department, Tsukiji-5
Chuo-ku, Tokyo, 104 Japan is responsible for the predictions and reductions of
timings of occultations of stars by the Moon. The International Occultation Timing
Association (IOTA) provides predictions of occultations.
9.41 Occultations of Stars
Whenever a star is within an angular distance from the Moon that is less than
the sum of the Moon's horizontal parallax and semidiameter, it will be occulted
from some point on the Earth. The large number of occultations make it impossible
to publish maps of each occultation in The Astronomical Almanac. It is possible,
however, to calculate and display these maps on a personal computer. The calcu-
lations require the Besselian elements, which are defined in a similar way to those
for eclipses. In occultations, the fundamental plane passes through the center of
the Earth and is perpendicular to the line joining the star and the center of the
Moon — i.e., to the axis of shadow. The origin of the coordinates is the center of the
Earth; the jc-axis is the intersection of the Earth's equator with the fundamental
plane and is taken as positive toward the east; the v-axis is perpendicular to that
of j: and is taken as positive toward the north in the fundamental plane. The great
distance of the star implies that the fundamental plane is perpendicular to the line
joining the center of the Earth to the star, and that the Moon's shadow is essentially
a cylinder whose intersection with the fundamental plane is a circle of invariable
size, its diameter being equal to that of the Moon. The coordinates of the center of
this circle — i.e., the axis of shadow — are denoted by x and y. The adopted unit of
linear measurement is the Earth's equatorial radius.
The Besselian elements are given for one instant only — namely, the time of
conjunction of the star and Moon in right ascension, when x is zero. They are:
To = the UT of conjunction in right ascension;
H = the Greenwich hour angle of the star at To',
Y = y at 7b;
x',y' = the hourly rates of change in x and y;
as, 6S = the right ascension and declination of the star.
9 / ASTRONOMICAL PHENOMENA 495
The formulas for x and y, the coordinates of the center of the shadow in the
fundamental plane, are
x sin 7T = cos 6 sin(a — as),
y sin 7T = sin o cos bs — cos c sin os cos(a — as),
where a and 6 are the right ascension and declination of the Moon at a particular
Universal Time and tt is the parallax.
For prediction purposes these reduce to
x=a~as cos 6, y= ^^+0.0087jc(a-af) sin^, (9.41-2)
7T 7T
where a, 6, and n are in degrees.
At conjunction
n 6 — 6,
x = 0, y=Y =
7T
, , , (9.41-3)
a cos 6 f>' it'
x' = , y'=°--Y-.
7T 7T
Here the primed quantities are hourly variations that arise from differentiating the
Moon's daily polynomial coefficients for a, 6, and n.
The rectangular coordinates of the center of the shadow on the surface of the
Earth at Universal Time f from conjunction are given by
x = x't,
y = y't+Y, (9.41-4)
where z is measured along the shadow axis from the fundamental plane toward the
Moon. The longitude and latitude (A, <j>) of this point on the surface of the Earth
are calculated from
4> = sin-1(ycos(5 + zsini5),
h - tan-1 7—^ -, (9.41-5)
— y sin 6 + z cos o
\ = h-(H+\Sx 1.0027380-
496 EXPLANATORY SUPPLEMENT
The rectangular coordinates of the edge of the shadow on the surface of the
Earth, (f , t), (}, are given by
£ = x — k sin Q,
r) = y-kcosQ, (9.41-6)
where k = 0. 2725 is the radius of the Moon in Earth radii and 0 cycles through the
range 0° to 360°. The longitude and latitude (A, <f>) of a point on the edge of the
shadow is obtained from Equation 9.41-5 by replacing (x, % z), with (£, jj, ().
The remainder of this section describes how to plot a map of the area of visibility
of an occultation. Project the longitudes and latitudes of the continents, islands,
and places of interest onto the plane that passes through the center of the Earth
and is perpendicular to the direction defined by a vector from the center of the
Earth to the place \c, <j>c. Here (Ac, 4>c) is the longitude and latitude of the center
of the shadow at the time of conjunction. This plane rotates with the Earth and
coincides with the fundamental plane at conjunction. The rectangular coordinates
of a place at longitude and latitude (A, <ft) are given by
u = cos <j> sin(A - Ac),
v = sin 4> cos 4>c — c°s 4> sin <f>c cos(A — \c), (9.41-7)
w = sin 4> sin <j>c + cos 4> cos <j>c cos(A — Ac),
where u and v are in the plane (with the origin at the center of the Earth), u is to
the east, v is to the north, and w is the height of the place above the plane; w > 0
for all places to be plotted.
To plot the edge of the shadow at any time during the occultation, calcu-
late the rectangular coordinates of the center of the shadow xq, yo, from Equa-
tion 9.41-4. Then calculate the coordinates £, rj, ( of the edge of the shadow from
Equation 9.41-6 and obtain the corresponding (A, from Equation 9.41-5. Finally
project the (A, 4>) OI the shadow outline onto the same plane as the continents, etc.,
using the coordinates (u, v) in Equation 9.41-7.
When £2 + rj2 > 1, the shadow is only partially on the Earth. To find the
values Q\ and Q2 where the edge of the shadow cuts the surface of the Earth, see
Figure 9.41.1, where O is the center of the Earth and S is the center of the shadow.
9 / ASTRONOMICAL PHENOMENA
497
Figure 9.41.1
Fundamental plane showing path of occultation
The great circle arc AB between the polar angle 8\ and 62 is then plotted by
setting x = cos 9, y = sin 9, and z - 0, and using Equations 9.41-5 and 9.41-7. The
polar angles are found as follows: Calculate <2i and Q2 from
sin(6i + 5) = ^ 1 and Q2 = 180° - 0, 2S,
2kD
where D = J (x\ + y§) and S = tan" 1 (y0 I x0).
Then for / = 1 and 2
6 ~x0 -k sin Qh
Vi =)'o - k cos Qit
and 9i = tan"1 (17,- / £,), 0° < 9t < 360?
The total area of visibility of the occultation may also be plotted by repeating
the calculation for a series of times between the start and the end of the occultation
on the same diagram.
A similar procedure may be applied for predicting the areas of visibility of
occultations of planets by the Moon.
(9.41-8)
(9.41-9)
498
9.42 Occultations of Planets
EXPLANATORY SUPPLEMENT
The prediction of occultation of stars (or minor planets) by planets (or by minor
planets) largely centers on the search for conjunction in right ascension within the
limits of difference of declination that make an occultation possible. These limits
are very small, being approximately the sum of the horizontal parallax and the
semidiameter of the planet.
The observations themselves yield useful information on the diameters of plan-
ets and about the shape of minor planets.
The actual prediction follows the basic principles underlying those for eclipses,
occultations, and transits; but more direct and less formal methods are used because
the angles involved are much smaller and the prediction much less precise.
An occultation will take place as seen from some point on the Earth's surface,
provided the difference 8 — 6S in apparent declination at the time of conjunction in
right ascension satisfies the condition
\6 - 6S\ < (tt + s)/ \smp\, (9.42-1)
where ir and s are respectively the equatorial horizontal parallax and semidiameter
of the planet and p is the position angle of its direction of motion. Note must be
taken of the position of the Sun, as for lunar occultation, because it is generally
useless to predict occultations of stars in daylight.
Methods of prediction are described by Taylor (1955); the paper also described
the determination of the actual limits of occultation. The IOTA (see Section 9.4)
provides predictions of planetary occultations that are published in Sky and Tele-
scope and the Astronomical Journal. The predictions of planetary appulses and
occultations are also published in The Handbook of the British Astronomical Soci-
ety.
9.5 POLE-STAR TABLES
The proximity of the second-magnitude star a Ursag Minoris, (Polaris or the Pole
Star), to the north pole of the sky has given it a special significance for the con-
venient determination of direction and latitude. This is particularly so in the fields
of navigation and surveying, for which the Pole Star's constant availability for ob-
servations (in northern latitudes) and the simple methods that can be used for the
reduction of observations are invaluable. For the more precise requirements of as-
tronomy, the Pole Star's distance from the true pole is sufficiently large that the
special methods of reduction no longer confer an advantage over standard meth-
ods. Thus Pole-Star tables are restricted to the precision required in navigation and
surveying. The principal table with the precision required by surveyors is included
in The Astronomical Almanac, pages B64-B67.
9 / ASTRONOMICAL PHENOMENA 499
Zenith
Figure 9.5.1
PZS triangle for Polaris
An alternative to the Polaris table is to use expressions involving coefficients
for the GHA and the polar distance p (see Section 9.51). These coefficients are
provided on page B63 and include the 5.5 magnitude star o Octantis, which lies
close to the south celestial pole.
If the polar distance of Polaris is denoted by p (of the order of one degree) and
its local hour angle by t, then its altitude h and azimuth Z, as seen from an observer
in latitude (f>, are given by solving the spherical triangle (see Figure 9.5.1) formed
by the North Pole, the zenith, and Polaris, S.
Because p is small, the solution may be expanded as
h = <f>+p cos t — —p sin p sin2 f tan 4> + ■■■ ,
2 (9.5-1)
Z cos <j> = p sin f + p sin/? sin t cos t tan $ + ■■■.
In each case, the next term of the expansion is of order />sin2p tan2 0 and cannot,
for many years, exceed O'.l for latitudes up to 70°.
EXPLANATORY SUPPLEMENT
For convenience of tabulation, these expressions are rewritten in the form
<f> - h = - Oocos/o - -posinpo sin2fotan0o)
+ - p0 sinpo sin" /0(tan <t> - tan <j>o)
- (p cos t - po cos t0) -ao + ai+a2, (9 5 ~
Z cos 0 = - (p0 sin t0 + po sinp0 sin r0 cos r0 tan fa)
- po sin pa sin /0 cos fo(tan <p - tan <fo)
- (p sin / - po sin t0) = b0 + b] + b2,
in which p0 and ?0 are the polar distance and hour angle of a convenient point close
to the mean position of Polaris throughout the year, and cj>0 is a mean latitude,
usually chosen to be 50°. The mean position of Polaris (which must not be confused
with its mean place) is usually chosen to have convenient exact values for its right
ascension cto and polar distance p0.
The first terms (flo,&o) in the modified expressions are functions of a single
variable, local sidereal time, LST, because
t0 = LST - q0 (9.5-3)
and may be tabulated at a suitable interval of LST.
The second terms (oi,f>i) are functions of t0 (i.e., of LST) and of latitude and
must thus be tabulated in a double-entry table with arguments LST and latitude.
By incorporating a mean value (corresponding to latitude 0O) in the first term,
the magnitude of these terms can be kept down to about 0!5; terms may thus be
tabulated at wide intervals of latitude and LST.
Similarly, the third terms (o2, b2) are functions of r0 and of the apparent position
of Polaris (i.e., of date). By proper choice ofp0 and o0, the magnitude of these terms
can be kept down, during the year, to about 0'.5; and they can also be tabulated at
wide intervals of date and LST.
As will be seen from the Polaris table in The Astronomical Almanac, the single-
entry table of o0 and b0 is arranged in twenty four columns, each containing values
for one hour of LST; this arrangement enables separate tables of a\ and b\, and
of a2 and b2, to be given for each hour of LST. In the column corresponding to
the hour of LST all these terms are taken from single-entry tables — the first with
argument minutes and seconds of LST, the second with argument latitude, and
the third with argument date. The error in using the tables for the hour, without
interpolation for LST, is greatest for the second term b\ (owing to its dependence
on sin 2 to) and may reach 0'.15 for extreme latitudes; otherwise the error is small.
9 / ASTRONOMICAL PHENOMENA 501
The complications of these tabulations are unnecessary for astronomical use
but valuable for navigational use, in which simplicity of tabular entry and of inter-
polation are of foremost importance. The Polaris table in The Nautical Almanac
is essentially the same as the table in The Astronomical Almanac, except for the
intervals of tabulation (1° in LST or LHA of the first point of Aries) and from a
further simplification for the user by adding constants (whose sum is one degree) to
AO, flj, a2 to make them always positive. Since lower precision is required in azimuth
the b coefficients have been replaced by a simpler table of azimuth as a function of
LHA Aries and latitude.
9.51 Derivation of the Pole Star Coefficients
The Greenwich hour angle (GHA) and polar distance (p) are expressed as series of
polynomial terms that allow for precession and proper motion, and also as series
of trigonometric terms that contain the leading terms in aberration and nutation.
Thus
GHA = ao +a[L + a2 sinl + 03 cosL + 04 sin M + o5 cos M
+ as sin 2L + a-j cos 2L + 1 5UT,
(9.51-1)
p = «o + a \L + a2 sin Z. + 0.1 cosL + 04 s'mM + o5 cos /V/
+ oe sin 2L + 07 cos 2L,
where Oo, «i. -.,07 are two sets of constants in degrees, UT is the Universal Time
in hours, L = L'n, M = Wn (where L' and J?' are the rates of change of the mean
aberrated longitude of the Sun and the mean longitude of the ascending node of the
Moon), and n is the interval from the beginning of the period to the time required.
The coefficients are determined by the method of least squares. Annual coef-
ficients are derived by taking daily values. Using all eight coefficients, the annual
fit gives maximum errors of ± 0"2. If only the first four terms are used in the
least-squares approximation then the maximum errors increase to ± l'.'O. Pole Star
coefficients have been calculated for the ten-year period 2096 to 2105, based on its
FK4 position and proper motion. The north polar distance is found to decrease from
its present value of about 48' to 27' 10" on 2100 March 20 and will then increase.
The distance will reach 1° by about 2250 and 2° by about 2450. (See Figure 9.51.1.)
The Pole-Star coefficients may be used to calculate the UT of upper culmination
of Polaris (UC) at the observer's longitude, the UT of the elongation west (EW),
and the elongation east (EE) at the observer's longitude and latitude.
502
EXPLANATORY SUPPLEMENT
2100
2104
Year
Figure 9.51.1
Polaris north polar distance 2096-2105
Calculate GHA0, and pQ at 0" UT. Then calculate the local hour angle LHAo -
GHAq + A, where A is the observer's longitude. Upper culmination occurs at time
fh _ 360 - LHAo
15.041069
(9.51-2)
and also at time /h — 23. 934 on the previous day. Elongation east and elongation
west occur at times
cos""1 (tan i^tanpo)
f ±
15.041069
(9.51-3)
where the plus sign is used for EW, and the minus sign for EE, and <j> is the observer's
latitude. If necessary add or subtract 23*934 to the time of the phenomenon to
obtain a time nearer the time of observation.
9.6 REFERENCES
Bennett, G.G. (1982). "The Calculation of Astronomical Refraction in Marine Navigation"
J. Inst. Nav. 35, 255.
Janiczek, P.M. and DeYoung, J. A. (1987). "Computer Programs for Sun and Moon Il-
luminance with Contingent Tables and Diagrams" United States Naval Observatory
Circular 171 (U.S. Naval Observatory, Washington).
Radio Corporation of America (1974). "Sources of Radiation" Electro- Optics Handbook
Section 6, 61.
9 / ASTRONOMICAL PHENOMENA 5Q3
Taylor, G.E. (1955). "The Prediction of Occupation of Stars by Minor Planets" J.B.A.A
65, 84.
Woolard, E.W. and Clemence, G.M. (1966). Spherical Astronomy 215 (Academic Press
New York).
CHAPTER
Stars and Stellar Systems
by D. Pascu, J.A. Mattei, C.E. Worley, H.C. Harris, J. Andersen, W.P. Bidelman,
J.H. Taylor, G. Lynga, R.E. White, G. de Vaucouleurs, TM. Heckman,
G,H. Kaplan, U.K. Pauliny-Toth, and J.F. Dolan
Section
Author
Afhliation
{10.1-10.15)
Anonymous
(10.2, 10.21)
D. Pascu
U.S. Naval Observatory
(10.22)
J.A. Mattei
Am. Assoc. Variable Star Observers
(10.23)
C.E. Worley
U.S. Naval Observatory
(10.24)
H.C. Harris
U.S. Naval Observatory
(10.25)
J. Andersen
Copenhagen Univ. Astron. Observatory
(10.26)
W.P. Bidelman
Case Western Reserve University
(10.27)
J.H. Taylor
Princeton University
(10.31)
G. Lynga.
Lund Observatory
(10.32)
R.E. White
University of Arizona
(10.33)
G. de Vaucouleurs
University of Texas
(10.34)
TM. Heckman
Johns Hopkins University
(10.41)
G.H. Kaplan
U.S. Naval Observatory
(10.42)
I.I.K. Pauliny-Toth
Max Planck Institute
(10.43)
J.F. Dolan
NASA/Goddard Space Flight Center
10.1 SOURCES OF DATA ON STARS AND STELLAR SYSTEMS
The data on stars and stellar systems are basically provided in catalogs. Positional
star catalogs can be divided into types. Observational catalogs provide the results
of observing programs with a given instrument. Compiled catalogs are compilations
of observational catalogs, which are usually assembled over an extended period of
time so that proper motions of stars can be determined as well as positions.
505
506 EXPLANATORY SUPPLEMENT
Observing programs and catalogs can be classified as fundamental or differen-
tial. In fundamental observing programs, the coordinates of the observations are
determined independently as part of the observing program. For example, the pole,
equator, and equinox might be determined from observations of circumpolar stars
and solar system objects. Differential observations are made with respect to star
positions from a reference catalog. All photographic and small field observations
are differential.
Special-purpose catalogs give a specific type of data, such as parallaxes, radial
velocities, and so on.
Heterogeneous catalogs are a collection of material from various sources. In
many cases, the catalogs are limited to a region of the sky, a range of magnitudes,
or a given type of object. The stars are identified by a catalog number or by a
constellation name and designation. A limited number of the brightest stars have
specific names.
10.1 1 Compiled Catalogs of Stellar Positions and Motions
Compiled catalogs are designated by a systematic reference system on the sky, which
is based on a selected list of stars included in the catalog. The positions and proper
motions for all the stars are on the same system of coordinates. Observational
data from a number of different instruments are included to minimize systematic
instrumental errors and to achieve the optimum accuracy. Compiled catalogs are
given on a standard equinox and epoch such as 1900.0, 1925.0, 1950.0, or 2000.0.
This is achieved by applying proper motion and precession to the compiled observed
positions made at different times. (Thus, the epoch of the catalog is not the average
date of the observational data in the catalogs.)
The compiled catalogs can be grouped into two series. The German series in-
cludes The Fundamental Catalog, The Neuer Fundamental- Catalog, and the FK3,
FK4, and FK5-Basic catalogs. The American series of catalogs includes Newcomb's
catalog of fundamental stars, Eichelberger's standard stars, the GC catalog, and
the N30 catalog. In addition, there is the SAO catalog, which is a heterogeneous
combination of a number of catalogs. The methods for converting the positions from
one epoch and reference system to another epoch and reference system is given in
Chapter 3. Currently in preparation are the Southern Reference Star system, the
International Reference Star system, and Part II of the FK5 catalog.
In the future it is expected that radio sources or quasars will be the reference
sources to which astronomical positions will be referred. These sources have the
advantage of lacking apparent motion due to their distance. Thus they provide a
type of inertial or space fixed reference system.
10 / STARS AND STELLAR SYSTEMS
10.12 Standard Reference Catalogs
507
Reference catalogs are of differing types. Some compiled reference catalogs are inde-
pendent sources of positions and proper motions. Examples are the FK3, FK4, and
FK5 catalogs. Other catalogs are dependent on the independent catalogs to deter-
mine their reference positions. Examples are the AGK2 and AGK3, which require
the FK catalog positions.
Observational catalogs are fundamental, or absolute, if the reference frame can
be determined from the observations themselves. The independent compiled cata-
logs must include some fundamental observational catalogs. They can also include
differential observational catalogs where the observations are made with respect to
some reference catalog.
Lists of star catalogs for the eighteenth and nineteenth centuries are given in the
volumes of Geschichte des Fixsternhimmels (Karlsruhe, 1922-1957; Berlin, 1952-
1959). A further list for the period 1900-1925 is given in Index der Sternorter 1900-
1925 (Bergedorf, 1928). Foremost among the observational catalogs in these lists
are those compiled under the auspices of the Astronomische Gesellschaft (A.G.) as
a cooperative effort by a number of observatories. This series of volumes was begun
in 1863, and gives the positions of all stars shown in the Bonner Durchmusterung
(B.D.) from decimation +80° to -18° to magnitude 9.0. The observations made
by each observatory were confined to a narrow zone of declinations best suited to
the latitude of the observatory. In more recent times photography has been used to
reobserve the A.G. zones. The new positions are given in:
AGK2 Zweiter Katalog der Astronomischen Gesellschaft fur das Aquinoktium
1950; ten volumes cover declination +90° to +20° (Hamburg-Bergedorf,
1951-1954), and five volumes cover decimations +20° to -T (Bonn, 1957-
1958).
AGK3 Zweiter Katalog der Astronomischen Gesellschaft fur das Aquinoktium
1973; based on plates taken at Bergedorf between 1959 and 1961. Proper
motions were formed using AGK2 as first-epoch plates. AGK3 R is the
catalog of reference stars for the AGK3.
Yale Catalogs of the zones +85° to +90°, +50° to +60°, -30° to +30°, and -60°
to -90° are published in Transactions of the Astronomical Observatory
of Yale University (New Haven, 1925 onward); the positions are also for
equinox 1950.0.
Among modern reference catalogs, the following are representative.
GC General Catalogue of 33342 Stars for the Epoch 1950, in 5 volumes, Wash-
ington, 1937. These catalogs give positions and proper motions of all stars
brighter than magnitude 7, and also include thousands of fainter stars.
508
EXPLANATORY SUPPLEMENT
FK3 Dritter Fundamentalkatalog des Berliner Astronomischen Jahrbuchs, I Teil
Veroffentlichungen des Astronomischen Rechen- Institute, no. 54, 1937. This
catalog gives positions for 1925.0 and 1950.0 of the 925 stars of Auwers'
Neue Fundamentalkatalog (1910). If Teil: Abhandlungen der Preussischen
Akademie der Wissenschaften, Phys.-Math. Klasse, no. 3, 1938. This cat-
alog gives the positions of 666 additional stars for equinox 1950.0.
FK4 Fourth Fundamental Catalog (FK4), Veroffentlichungen des Astronomis-
chen Rechen-Instituts, Heidelberg, no. 10, 1963. A revision of FK3; the
re-examination of the available observations showed that no change in the
equinox was justified.
FK5 Fifth Fundamental Catalog (FK5), Part 1: The Basic Fundamental Stars.
Veroffentlichungen Astronomisches Rechen-Institut, Heidelberg, no. 32, 1988
A revision of FK4; the re-examination of the available observations showed
that changes to the constant of precession and a correction to the equinox,
both at the epoch and as a function of time, were necessary. See Section 3.5
for the transformation between the FK4 and the FK5 system.
N30 "Catalog of 5268 standard stars, 1950.0, based on the normal system N30,"
Astronomical Papers of the American Ephemeris, 13, part III, 1952. The
positions are derived from more than 70 catalogs with epochs of observation
between 1917 and 1949.
ZC "Catalog of 3539 zodiacal stars for the equinox 1950.0," Astronomical Pa-
pers of the American Ephemeris, 10 part II, 1940. This catalog gives posi-
tions of all stars to magnitude 7 in the zodiacal zone as well as the positions
of many fainter stars. It is intended for use in occultation work. It is based
on 90 catalogs, and positions are reduced to the FK3 system.
PFKSZ "A preliminary general catalog of fundamental faint stars between de-
clinations +90° and -20°," M.S. Zverev and D.D. Polozhentsev, Publica-
tions of the Main Astronomical Observatory of Pulkovo, 72, 1958.
10.13 Observational Positional Catalogs
Observational data from a given instrument over a period of time are usually pub-
lished as a catalog. In this case, the data have been reduced in a systematic manner
and are published on some reference system, either on the system of a compiled
catalog or on a system as determined by the instrument. Because the method of
reduction and the reference system can vary for observational catalogs, the intro-
duction or description of the catalog must be reviewed carefully, and the use of
observational catalogs and the combining of observational catalogs must be done
with care. Observational catalogs usually combine a number of observations oi a
given star to determine the position of that star, and give a mean epoch for the
time of the observations. This epoch may differ from the epoch of the reference
system on which the stars are given. Examples of observational catalogs are the ob-
servations of the six-inch transit circle in Washington (which have been published
in the Publications of the U.S. Naval Observatory), Greenwich observations (which
10 / STARS AND STELLAR SYSTEMS 509
have been published in the Greenwich Observations series), Radio source lists, and
many other observations from other observatories.
10.14 Other Catalogs and Lists
Compiled catalogs and observational catalogs generally deal with positions and
proper motions of stars. In addition, there are catalogs of double stars and catalogs
giving the radial velocities, parallaxes, spectra, and other characteristics of stars.
In some cases, the catalog can be based on single observation of each star; in other
cases, such as parallaxes, many observations of the stars are necessary in order to
determine the parallaxes.
10.15 Data Centers and Their Facilities
With the introduction of data-processing equipment, organizations use the equip-
ment to develop a collection of data in machine-readable form. The data include
data currently being developed and also older data that was coded into machine-
readable form for analyses purposes. In general, these data collections were located
at the places actually making the observations or at the ones most dependent upon
computer analyses — for example, the U.S. Naval Observatory, Royal Greenwich
Observatory, Astronomisches Rechen-Institut, and Pulkova Observatory.
The number of different locations for collections of data proliferated, and the
need for central data centers became evident. Therefore, the Centre de Donnees
Stellaires was established in Strasbourg, France. Subsequently, a data center was
established at the NASA Goddard Space Flight Center. These two organizations
have become central depositories for data. They obtain copies of the data from
previously established collections. In addition, they cross-index and consolidate the
information from many different systems into a single file of data on each star.
Computer networks allow astronomers to directly access the data center to
obtain the most current information of the desired specific data. One system—
the Set of Indentifications, Measurements, and Bibliography for Astronomical Data
(SIMBAD) system^allows direct access to the computer at the Strasbourg data
center and the capability to obtain all the current information concerning a specific
star.
1 0.2 STELLAR DATA I N THE ASTRONOMICAL ALMANAC
The progenitors of The Astronomical Almanac (The Astronomical Ephemeris [AEJ,
formerly The Nautical Almanac, and The American Ephemeris and Nautical Alma-
nac [AENAJ) have always included stellar data in the form of lists of apparent
51 0 EXPLANATORY SUPPLEMENT
places and/or mean places of bright stars. Such lists were used by astronomers for
the positional calibration of their instruments and formed the basis for navigational
and surveying almanacs. The precision to which these lists were published varied
from Of 1 in right ascension and 1" in declination, to 0?001 in right ascension and
0!'01 in declination. With the publication of Apparent Places of Fundamental Stars
in 1941, the number of apparent places printed in the almanacs was greatly di-
minished, and by 1957, apparent places for stars were omitted altogether. In the
1960 editions, the AE and AENA were unified in content, but kept their separate
names. The purpose and scope of the almanacs had also evolved considerably. The
emphasis was no longer on the "practical requirements of navigators and survey-
ors," but on the "requirements of the practical astronomer" {Exp. Supp. pp. 10-11).
For stellar data, this resulted in mean places for stars brighter than 4.75 (1078) to
an accuracy of 0U in right ascension and 1" in declination. Visual magnitudes and
spectral types were also given, though they were not up-to-date.
With the 1981 edition, the unification of the almanacs was completed by the
adoption of a single printing with one title — The Astronomical Almanac. This edi-
tion was a complete revision in content as well as format. To make the almanac
more responsive to modern astronomical research, the scope of this new revision
was enlarged to include an improved and expanded stellar data section.
The general precepts for establishing the contents of the section were as follows.
Since The Astronomical Almanac was placed handily near most telescopes, finding
lists for the most frequently observed objects was considered most useful. These
stellar or nonstellar objects could be the subjects of investigations or they might be
used for calibrational purposes. Categories for these objects were identified (some
fell in several categories) and the data most pertinent to the observations were
listed.
Since the names of the astronomical objects researched were quoted in the lit-
erature, studies with the Bibliography of Astronomical Objects of SIMBAD (e.g.,
Ochsenbein and Dubois 1986) were indispensable in determining the most fre-
quently quoted names. Ochsenbein and Dubois have shown that the most popular
objects now are nonstellar, and include galaxies, QSOs, globular clusters, X-ray
sources, and pulsars. Calibrational objects were not as readily identified since they
were not usually cited. However, calibrational systems were always mentioned, and
studies such as that of Mermilliod (1984) were useful in determining which systems
were most frequently used. Ideally, lists recommended by the IAU, such as the list of
radial velocity standards (IAU Trans. 1957) were the most desirable. Unfortunately,
that was the only such list, and it was hopelessly out-of-date.
Once a selection of tables (lists) was made, internationally recognized experts
were contacted to help make the selection of the objects and data to be listed. These
experts, in general, had themselves published catalogs or compiled unpublished
data files of astronomical objects. Selections from these catalogs (files) were then
proposed for publication in The Astronomical Almanac. At present, the almanac
10 / STARS AND STELLAR SYSTEMS
511
offers fourteen lists in 13 tables. A mixture of frequently observed objects as well as
commonly used calibrational systems is included. Even in those lists not specifically
defining a calibrational system, much of the data can be used for calibrational
purposes. Titles and short descriptions of these tables are:
Bright Stars: 1482 bright stars, listing positional (FK5, SAO), photometric
(UBV horn Nicolet, 1978) and spectroscopic data (from W.P. Bidelman,
1991). Also notes on duplicity, variability, and reliability of data (from Yale
Bright Star Catalog, 4th ed., Hoffleit and Jaschek, 1982).
Bright MK Atlas Standards: Included in the Bright Stars list. Recent MK Atlas
Standards, which are included in the Yale Bright Star Catalog.
UBVRI Standards: 107 bright standards selected by H.L. Johnson.
Stroragren four-color and H-heta standards: 319 bright standards selected from
list of Perry et ah (1987)
Radial- Velocity Standards: Bright and faint IAU lists (IAU Trans. 1957, IAU
Trans. 1973). Under revision.
Selected Variable Stars: 181 bright variables of different types, selected by J. A.
Mattel from the fourth edition of the General Catalog of Variable Stars;
Sky Catalog 2000.0, Volume 2; and data files of the American Association
of Variable Stars Observers.
Selected Open Clusters: 288 open clusters selected by G. Lynga (1988) from his
computer-based Lund Catalog of Open Cluster Data.
Globular Clusters: Data on 157 galactic globular clusters compiled by R.E.
White.
Bright Galaxies: 194 of the brightest and largest galaxies, selected by G. de
Vaucouleurs from the Second Reference Catalogue of Bright Galaxies (1976).
Radio Source Positions: 233 radio sources suitable for positional calibration;
list of Argue et al. (1984).
Radio Flux Calibrators: 14 radio sources suitable for flux calibration, from Baars
et al. (1977), as updated by the authors.
Selected Identified X-ray Sources: 194 identified X-ray sources, selected by J.F.
Dolan from his unpublished survey file.
Selected Quasars: 99 QSOs selected by T.M. Heckman from A Catalog of Quasars
and Active Nuclei (Veron-Cetty and Veron, 1985).
Selected Pulsars: 92 Pulsars selected by J.H. Taylor from the compilation of
Manchester and Taylor (1981).
More information on these astronomical objects and tables, with references
to additional data, are given in the sections that follow. Although these tables
include about 90% of the 100 most frequently quoted objects in the literature,
there are some noticeable gaps, including a table of nebulae and a list of infrared
objects. It has also been suggested that fainter calibrational objects, such as fainter
photometric standards, be listed. Fainter calibrational objects will be included in
future editions, but the problem of finding charts has not yet been resolved. The
512 EXPLANATORY SUPPLEMENT
form and content of the section are incomplete and experimental; it is expected to
change continually to reflect the changing needs of astronomical research.
10.21 Bright Stars
Tables of bright stars have been published in the national almanacs from their
inceptions. This affirms the continuing usefulness of bright-star data. The positions
of the stars are the most important data to users. These data are used in navigation
and topographical surveying and by observational astronomers for the positional
calibration of telescopes and instruments.
At present the Yale Bright Star Catalog (Hoffleit and Jaschek, 1982) is the
most comprehensive hardcopy collection of data for stars brighter than 6.5 visual.
It contains astrometric, photometric, and spectroscopic data as well as duplicity
and variability information for the 9110 stars of the Harvard Revised Photometry.
The Sky Catalogue 2000.0, Volume 1 (Hirshfeld and Sinnott, 1982), is less com-
prehensive than the YBSC, but contains information on 45,269 stars brighter than
8.05. Neither tome, however, can compare with the computer-based astronomical
data bank of the Centre de Donnees Stellaires at Strasbourg Observatory called
SIMBAD (Set of Identification, Measurements, and Bibliography for Astronomical
Data). This data bank gives, for each of 600,000 stars, an exhaustive array of data
as well as citations for each star in bibliographical references to papers in 90 ma-
jor astronomical journals since 1950 (Egret and Wegner, 1988; Wegner, Egret, and
Ochsenbein, 1989).
When work began on the Bright Stars list in the late 1970s, only the third
edition of the YBSC was available, and most of its data had been superseded. For-
tunately, some of the more important new data were already available in computer-
readable form. A selection of less than 1500 stars (30 pages) was made from the
third edition: all stars brighter that 4.5 visual were included. In addition, FK4 stars
brighter that 5.5 visual were included to insure that the positions of fainter stars
were precise to 1 arcsecond. Finally, a short list of bright MK spectroscopic stan-
dards were incorporated in the list to avoid unnecessary duplication. These were
YBSC stars that were in the atlas by Morgan, Abt, and Tapscott (1978), and the
atlas of Keenan and McNeil (1976).
Following the IAU-recommended practice of giving two identifications for each
object, the Bayer/Flamsteed designations were selected because of their wide-spread
use, and the BS (HR) number to facilitate access to additional data given in the
YBSC. The positions of the stars are still the most important data to the observer,
and for most applications, a precision of one arcsecond is sufficient. To avoid listing
the proper motions, yet insure the stated precision for the whole year, the positions
are brought to the mean equator and equinox of the middle of the current year,
with proper motions applied. Reduction to apparent places may be performed using
the methods given in Section B. Starting positions for the stars were taken from
10 / STARS AND STELLAR SYSTEMS
513
the FK5, GC, or SAO catalogs— in that order of preference — and all positions were
reduced to the FK5 system.
Photometric quantities on the UBV system were obtained from Nicolet (1978)
except for some double stars for which Nicolet lists the combined magnitudes and
color indexes. In these cases, the magnitudes and color indexes were taken from the
third edition of the YBSC. Spectral types on the MK system (if available) or the Mt.
Wilson system have been kept current by W.P. Bidelman (1991) from his extensive
card catalog. Last, but not least, the notes give the duplicity and variability status
of each star as well as indications of the reliability of the data. It is in these notes
that the YBSC (fourth edition) has been most useful.
The annual publication of The Astronomical Almanac presents a considerable
opportunity for the Bright Stars list to offer the observer the most current and
relevant data for the brightest stars.
10.22 Variable Stars
The information in the list of variable stars has been compiled using the references
in the third edition of the General Catalogue of Variable Stars and its three sup-
plements; the Sky Catalogue 2000.0, Volume 2; and the data files of the American
Association of Variable Star Observers (AAVSO). Only bright stars from each class
of variable stars that also fit the criterion of having an amplitude of variation 0.5
magnitude or greater have been selected. The limiting magnitude for each type of
star selected is indicated in The Astronomical Almanac.
Complete, up-to-date information on about 28,450 variable stars discovered
and named by 1982, and on the constellations Andromeda through Vulpecula is
given in the fourth edition of the General Catalog of Variable Stars (Kholopov
et ai, 1957-1987) (GCVS), published in three volumes from 1985 to 1987 by the
Astronomical Council of the USSR Academy of Sciences in Moscow. This catalog
contains the 1950.0 position, precession constants, galactic coordinates, reference
and finder chart information, type, range of variation, epoch for the elements given,
period, and spectrum for each star. Each volume also contains a section of remarks
on stars that require them. Two more volumes will be published as part of this
edition, and will contain positional information for the equator and equinox of
1900.0 and a list of variable stars in order of right ascension.
The information in the fourth edition of the GCVS is also on magnetic tape,
and may be be obtained from the Astronomical Data Center (Mail Code 633, NASA
Goddard Space Flight Center, Greenbelt, MD 20771, U.S.A). Bound copies of the
first three volumes are also available from the AAVSO, 25 Birch Street, Cambridge,
MA 02138, U.S.A.
Stars that have been discovered and named from 1982 to 1988 are given in the
67th, 68th, and 69th Name Lists of Variable Stars, published by Commission 27 of
514 EXPLANATORY SUPPLEMENT
the IAU as Information Bulletin of Variable Stars (Numbers 2681, 3058 and 3323,
respectively).
The nomenclature used in the list of variable stars in both The Astronomical
Almanac and the General Catalog of Variable Stars is the conventional method
developed by F.W.A. Argelander in the mid-1800s. According to this method, the
first star discovered in a constellation is given the letter R, followed by the genitive
form of the Latin name of the constellation. Subsequent variables in the same
constellation are named S, T, and so on to Z, followed by RR to RZ; then SS to SZ,
and so on to ZZ, after which follow AA to AZ, and so on to QZ, the 334th variable
in a constellation. The next variable found in a constellation is named V335 plus
the genitive. Stars which have been assigned Greek letters prior to the start of this
system, such as Omicron Ceti, or small Roman letters, such as g Herculis, continue
to keep those names.
Extensive references on variable stars may be obtained through the interna-
tional database SIMBAD maintained by the Strasbourg Centre de Donnees Stel-
laires (CDS) (11, rue de l'Universite, F-67000 Strasbourg, France).
The AAVSO is the largest organization of its kind — compiling observations
on variable stars from observers worldwide. These observations are digitized and
processed and kept up-to-date. Since its inception is 1911, the AAVSO has the
largest database of visual variable-star observations, including data on about 3000
variable stars having an amplitude of variation 1 magnitude or larger. The AAVSO
also has a database of photoelectric observations on 50 small-amplitude (smaller
than one magnitude) red variable stars. Copies of both visual and photoelectric
observations may be obtained from the AAVSO on request by writing to Dr. Janet
A. Mattei, Director, AAVSO, 25 Birch St., Cambridge, MA 02138.
Other locations that act as repositories of photoelectric observations, together
with the contact person, are:
Dr. C. Jaschek
Centre de Donnees Stellaires
Observatoire de Strasbourg
11, rue de l'Universite
F-67000 Strasbourg
France
Mrs. E. Lake, Librarian
Royal Astronomical Society
Burlington House
London, WIV ONL
Great Britain
10 / STARS AND STELLAR SYSTEMS
515
Dr. E. Makarenko
Odessa Astronomical Observatory
Shevchenko Park
Odessa 270014
USSR
1 0.23 Double and Multiple Stars
The structure and evolution of a star is basically dependent on two parameters: (1)
the initial chemical content, and (2) the initial mass. Double-star systems permit the
determination of the latter parameter, hence their importance in astrophysics. Their
numbers and distribution also provide insights into the process of star formation.
Double and multiple stars form an important fraction of the stellar population.
The term double star includes the optical pairs formed by chance juxtaposition of
unrelated stars as well as the true binaries that are physically related. For many
double stars there is not enough information to determine whether the pair is phys-
ical; therefore catalogs also include optical doubles. Historically, double stars have
also been differentiated by the observational means used to discover and study
them. Thus, there are not only visual doubles, but also spectroscopic, astromet-
ric, photometric, interferometric, and occupation pairs. With the development of
modern observing techniques, however, these distinctions are likely increasingly to
become blurred.
Multiple systems, containing more than two components, generally are classed
as either "hierarchical" or "nonhierarchical." In the first instance, a typical triple
will contain a close pair and a third component located at a distance many times
the separation of the close pair. In the second instance, the separations of all the
components are roughly the same. Examples of these classes are Epsilon Hydrae
(hierarchical) and the Trapezium (nonhierarchical).
Nomenclature for double stars is extremely mixed. Visual pairs are usually
designated by the discoverer's abbreviation and serial number, although some ob-
servers have neglected to number all or a portion of their pairs. Eclipsing pairs
are designated by the variable star nomenclature. There appears to be no standard
nomenclature whatsoever for spectroscopic, astrometric, or occupation binaries,
which are given a multitude of designations taken from various catalogs.
With few exceptions, the primary (A component) of a visual double is the
brighter star; the secondary (B component) is fainter. There are systems of very
small magnitude difference, however, where this distinction is arbitrary. Additional
components in a multiple system are designated C, D, and so on, in order of mcreas-
ing separation. It sometimes happens that a new, close, companion is discovered in
an already defined system. The convention is then to designate it as Aa, Bb, Cc,
etc.
516 EXPLANATORY SUPPLEMENT
The Washington Double Star Catalog, 1984-0 (WDS) was issued in tape form.
It is the replacement for the Index Catalog of Visual Double Stars, 1961.0, issued
by the Lick Observatory, and commonly called the IDS. The format of the WDS
is nearly identical to that of the IDS; both are cardimage. The WDS tape includes
an introduction, the list of known double stars (numbering 73610), notes, and ref-
erences. The data file is being kept up-to-date, and a new tape edition should be
available in 1992. Currently, the WDS is ordered by .12000.0 coordinates, and cer-
tain other format changes have been made to modernize the data base. Users can
obtain tape copies of the WDS from either the Astronomical Data Center or the
Centre de Donnees Stellaires (addresses for these are given in Section 10.22).
The companion catalog to the WDS is the Catalog of Observations. This catalog
contains the differential measures of each pair. At the present time, the database
contains about 426,000 measures (means). It is complete for published observations
since 1927, and largely so for those of earlier date, which continue to be incor-
porated. This database is in a continual state of change, and consequently it is
not (and never has been) intended to be distributed as an entity. However, an at-
tempt is made to provide requestors with data on individual objects, provided that,
"reasonable" amounts of data are requested.
Eclipsing binaries are listed by Wood et al. (1980). Their list contains informa-
tion on more than 3500 systems, and includes a small number of objects that, while
interacting and producing light variations, are not strictly eclipsing systems. The
magnitude limit is 15 at minimum light. The left-hand page for each entry contains
identifications and approximate coordinates, maximum magnitude, depth of pri-
mary and secondary minima, spectral types of each component, epoch of primary
eclipse, and period. The right-hand page contains brief notes and references.
Spectroscopic binaries with orbits are listed by Batten et al. (1989). For the
1469 systems included, the authors list on the left-hand page the identification and
position, magnitude, spectral type, and reference. The right-hand page lists the
spectroscopic elements and a quality rating for each solution. Following the entries
are very extensive notes and references concerning individual systems. A somewhat
parallel effort by Pedoussaut et al. (1988) has resulted in many supplementary
catalogs, which are valuable for their attempt to list all bibliographical references
to each object. Unfortunately, there is no modern catalog for known or suspected
spectroscopic binaries that have no calculated orbit.
10.24 Photometric Standards
Tables of photometric standard stars are listed with their coordinates and with their
photometric standard values taken from the cited literature. Photometric standard
stars are widely used by observers because it is difficult to establish absolute cali-
brations of the telescopes and detectors used for stellar photometry and to account
10 / STARS AND STELLAR SYSTEMS
517
for atmospheric absorption. Standard values refer to intensities outside the atmo-
sphere. They are defined and listed using the magnitude system traditionally used
by astronomers. The magnitude scale is logarithmic and is defined as (-2. 5 log
Intensity). .
Many different photometric systems exist. They are defined using filters and
detectors that give different passbands. The first system, presently included in The
Astronomical Almanac is the system most widely used by astronomers. The UBV
system, introduced by Johnson and Morgan (1953), was the first to be used ex-
tensively. It has broad passbands (700-1000 A wide) with effective wavelengths of
3650 A, 4400 A, and 5500 A for U, B, and V, respectively. It is described by Johnson
(1963).' Later, R and / filters were added to the system by Johnson with effective
wavelengths of 7000 A and 9000 A. However, several variations on RI systems have
been developed and used by others. The RI systems most widely used are those of
Johnson, of Kron et al. (1957) and of Cousins (1976). The list of bright UBVRI
standards published in The Astronomical Almanac was selected by H.L. Johnson
from the list of Johnson et al. (1966). An extensive list of faint UB V standards was
published by Landolt (1973). Landolt's standards were located in the equatorial Se-
lected Areas, convenient to astronomers worldwide. A decade later, Landolt (1983)
published a list of 223 faint UBVRI standards. These stars were also located in the
same equatorial Selected Areas and included many of the stars in the first list. The
t/BVdata were calibrated by the earlier list while the RI data were tied to the RI
standards of Cousins (1976). r
The second system presently included in The Astronomical Almanac is the
widely used system of four-color and H-beta photometry established by Stromgren
(1963) and others. The passbands are of intermediate width (180-300 A) centered
at 3500 A, 4110 A, 4670 A, and 5470 A for «, v, b, and y, respectively, with a pan oi
passbands at 4680 A to measure the hydrogen Balmer line. In the 1981 through the
1990 editions of The Astronomical Almanac, the complete list of Stromgren four-
color standards, included in the Crawford and Barnes (1970) list, was published
along with beta indexes from Crawford and Mander (1966). Beginning with the
1991 edition, the new list of Perry, Olsen, and Crawford (1987) was used. Al 319
four-color standards are listed along with their beta indexes. The new list contains
several fainter stars (V = 6-7 mag.) as well providing more accurate data.
The tables in The Astronomical Almanac are provided for convenience, but
observers should keep in mind that the tables may not always provide a complete
list of standards. One requirement for a well-defined photometric system is that it
have a high degree of internal consistency. The tables included in The Astronomzcal
Almanac are chosen from sources with an effort to provide consistent data Ut
course, no set of data is perfect, and a sample of stars selected from a larger set ot
standards cannot completely define the standard system. Furthermore observers
should keep in mind the differences between variants of a system, such as the several
RI systems mentioned previously, and should seek to match their own passbands as
518 EXPLANATORY SUPPLEMENT
closely as possible to the system passbands. Data taken in one system can sometimes
be transformed to another system as described by Bessell (1986), Bessell and Weis
(1987), and Taylor (1986) for example. However, no general transformation exists
that can be applied to a wide range of stellar types and that is satisfactory at the
1% level of precision that is desirable for many purposes.
As large telescopes with more sensitive detectors come into common use, fainter
stars are being established as secondary standards in many photometric systems.
The needs of astronomers and the capabilities of their instruments will continue to
require changes in the stars used as standards.
10.25 Radial-Velocity Standards
The list of radial-velocity standard stars given in The Astronomical Almanac and
their standard velocities are a subset of those adopted by IAU Commission 30
(Trans. IAU, IX, 442, 1957; Trans. IAU, XVA, 409, 1973). Stars that have since
been proved to be definite velocity variables have been deleted, and a few velocities
that are now known to differ from their IAU value by more than lkms"1 have
been marked. Candidates for future adoption as primary standard stars (see below)
are also identified. V magnitudes are from Nicolet (1978) when possible; otherwise,
they are estimated from photographic magnitudes and spectral types. The latter
are from the Bright Stars list (pp. H2-H31), the Yale Bright Star Catalog, or the
original IAU list, in order of preference. Positions are obtained by the procedures
used for the Bright Stars list.
Some discrimination is necessary in the use of these standard stars, for the
following reasons.
Radial-velocity standards serve two basic purposes: (1) as constant-velocity
st ars with which to monitor instrument long-term stability, and (2) as stars of known
radial velocity with which to test instrument zero-points. Over the last decade or
so, the precision and efficiency of radiahvelocity measurements have improved by
factors of 10 to 100 or more. A dramatic increase in research based on radial- velocity
data has followed, much of the stellar work relying on high-precision radial velocities
(e.g., cluster dynamics, stellar pulsations, or low-mass companions).
It was realized some time ago that the old IAU standard system was inad-
equate to meet these increased demands (Batten 1985). Therefore, in 1985, IAU
Commission 30 created a Working Group whose goal was proposing a new and
better system, in which the zero-point and individual velocities would be known to
approximately + 100 ms-1. The plans for the new system, and the results obtained
during 1985-1988, are described in the reports of IAU Commission 30 (IAU Trans.
1988 and IAU Trans. 1990). Unfortunately, no new system has yet been adopted.
The reasons for this are relevant to the use of the present standard stars, and are
described in the paragraphs that follow.
10 / STARS AND STELLAR SYSTEMS 519
During 1985-1988, radial-velocity variations of tens to hundreds of ms-1 and
with periods ranging from days to years, have been shown to occur in solar and
later-type stars, both dwarfs (Campbell et al., 1988) and giants (Irwin et al., 1989).
However, few stars have been thoroughly examined at this level of precision. More-
over, since the low-amplitude variations may have very long periods, the constancy
of a given star to, say, ± 100 ms-1 cannot be established quickly. Finally, detailed
comparisons have revealed subtle systematic differences between the main instru-
ments now in use. These must be understood before definitive velocities for indi-
vidual stars can be adopted.
The longest possible history of precise observations was the main factor in se-
lecting candidates for a new standard system. Hence, potential new standards were
chosen from existing standards within ± 20° of the equator, making them observable
from both hemispheres, which eliminates proven variables. The candidates are indi-
cated in the list given in The Astronomical Almanac. Further observations of these
stars are being made, and are strongly encouraged, to identify those that are suffi-
ciently constant for later adoption as primary standards, and to define individual
mean radial velocities for them.
To avoid a proliferation of different velocity systems, the old IAU system will
remain in force until the new system can be adopted, even though the velocities of
some stars are known to differ from the IAU standard values listed. Observations
of minor planets have shown that the mean zero-point of the old system is correct
to within approximately ± 300 ms-1. Therefore, observations of several primary
standard candidates should yield a fairly good mean value; reliance on one or two
stars only is discouraged.
Finally, since radial-velocity observations of early-type stars are inherently less
precise than of solar and later-type stars, and because accurate, independent veloc-
ity checks are not readily available, satisfactory early-type radial- velocity standards
do not exist. Efforts to establish such standards are under way, but will take years
to complete. The present list is not suitable for checking the zero-point of radial-
velocity data for early-type stars.
10.26 Spectral Classification
A few words regarding spectral classification may be helpful to users of the Bright
Stars list. A star's spectral type is an alphanumeric notation that, when appro-
priately calibrated, can give quite reliable information on the absolute magnitude,
surface temperature, and chemical composition of the outer layers of a star. This
notation generally involves a "spectral class" ranging from O through M in or-
der of decreasing surface temperature, and a "luminosity class" ranging from la
through Vb in order of decreasing intrinsic brightness. A few exceptionally bright
supergiants are assigned luminosity class zero. Further, if a star's spectrum appears
to show a significant deviation in chemical composition from that of the Sun, an
520 EXPLANATORY SUPPLEMENT
indication of this abnormality is added. For the peculiar A stars, this involves spec-
ifying the elements seen in abnormal strength (Si, Mn, Eu, etc.). The stars with an
abnormal heavy-elements/hydrogen ratio are often indicated by use of an element
abbreviation (Fe, Ca, etc.) followed by a positive or negative number denoting the
degree of enhancement or weakness of the line of that element as seen in the spec-
trum. Stars of markedly unusual composition, such as the carbon stars (C) and the
very-heavy-element stars (S), as well as the Wolf-Rayet and white-dwarf stars, all
have their own schemes of classification.
Spectral classification notation can be a bit confusing — "n" or "s" indicates
broad (rotation) or sharp lines respectively, and "e" indicates hydrogen emission.
A "+" or "-" following the spectral class indicates that a star is very slightly later
(cooler) or earlier (hotter) than the class given, while a "+" or "— " following the
luminosity class indicates that the star is very slightly brighter or fainter than the
class assigned. An "m" indicates that the object is a metallic-line star, and a "e,"
"g," or "d" before the spectral class indicates that the object is a supergiant, giant,
or dwarf but has no actual MK classification.
The Morgan-Keenan system (MK) is defined by standard stars, of universally
accepted spectral type. Classifiers attempt to locate their unknowns within the mul-
tidimensional array defined by these standard stars. Unfortunately the results — at
least as exemplified in the preceding data — is not entirely satisfactory, for several
reasons. In the first place, the types assigned to most of the standard stars have
undergone considerable change during the past half-century, so that a given investi-
gator's determinations may be a function of time. Second, the instrumentation used
by the various spectral-type practitioners has varied considerably, with naturally
a very large attendant variation in the accuracy of the classification attempted,
if not actually realized. For example, most of the types listed are from rather in-
homogenous slit spectra, but some are from moderate-dispersion objective-prism
plates. Many of the types were determined long ago; others are very new. They are
certainly in no sense homogeneous.
In view of these situations, the best one can now do is to list what appear to be
the best current spectral types. Even the present list will be somewhat out-of-date
when it appears.
The calibration of Morgan-Keenan spectral types in terms of absolute magni-
tude, intrinsic color, and surface temperature has also been subject to considerable
change over time. The most useful recent reference is undoubtedly the extensive
work of Schmidt-Kaler (1982). Calibration of the third (abundance) parameter of
the system is currently in a primitive state, partly because there is not yet complete
agreement among the specialists in the field as to details of technique and notation.
It should be clear that even an accurate spectral type is no substitute for a
completely reliable spectral analysis and absolute-magnitude determination. But in
the stellar astronomers' real- world, spectral types have proved, and still are proving,
extremely useful in enabling us to obtain quickly a good, albeit necessarily rough,
10 / STABS AND STELLAR SYSTEMS 521
idea of the nature of a star, allowing us to group objects of similar properties
together for statistical treatment; and providing important and interesting objects
for astrophysical study.
The road from Secchi to Morgan and Keenan and beyond has been long and
winding, and its end is still now only faintly glimpsed.
Those who wish to learn more about the development and future trends of
spectral classification should find the following references useful:
R.H. Curtiss (1932)
W.W. Morgan, P.C. Keenan, and E. Kellman (1943)
P.C. Keenan and W.W. Morgan (1951)
W.P. Bidelman (1969)
W.W. Morgan and P.C. Keenan (1973)
P.C. Keenan and R.C. McNeil (1976)
W.W. Morgan, H.A. Abt, and J.W. Tapscott (1978)
M.F. McCarthy, A.G.D. Philip, and G.V. Coyne (1979)
T. Schmidt-Kaler (1982)
R.F. Garrison (1984)
R.O. Gray (1989)
10.27 Pulsars
Pulsars are selected for The Astronomical Almanac list if their mean 400 MHz flux
density is greater than 40 rnJy. Another handful are included that are particularly
interesting — for example, millisecond pulsars or members of binary systems. Most
of the data are taken from a compilation by Manchester and Taylor (1981), which
contains references to the original literature. Some recent updates and additions
are from a new compilation now being readied for publication (Lyne, Manchester,
and Taylor).
Pulsars are conventionally designated by their positions in the equatorial coor-
dinate system, equinox and equator of 1950.0, given as right ascension in hours and
minutes and declination in degrees (or, in case of ambiguity, degrees and tenths).
In most cases, the positions are measured by means of pulse-timing observations.
Arrival times measured on a number of dates, spread over many months or longer,
can provide astrometric accuracies well under a second of arc for "normal" pulsars,
and at, or below, the milli-arcsec range for millisecond pulsars (Rawley, Taylor,
and Davis 1988). The same observations yield high-precision values for the pulsar
periods, P, and period derivatives or spin-down rates, P.
The group velocity for meter-wavelength signals traveling through the interstel-
lar medium is less than c (the speed of light in vacuum), because of the presence of
an ionized gas component. The velocity is also frequency dependent, according to
the relation
v = c(l -v2plv2)m, (10.27-1)
522 EXPLANATORY SUPPLEMENT
where vp is the plasma frequency and v is the frequency of the wave (see, for ex-
ample, Manchester and Taylor, 1977, Chapter 7). For a given pulsar this effect is
directly measurable from timing observations at separated frequencies. It is quan-
tified in a constant called the dispersion measure, which amounts to the integral of
free electron density along the line of sight. The conventional units are cm-3 pc.
Irregularities in the interstellar medium, together with the extremely small
effective angular diameters of pulsars, cause another propagation effect not generally
seen in radio observations of other types of sources. Interstellar scintillation, the
result of multi-path propagation through an irregular medium, can cause larger
(order-of-magnitude) variations in apparent flux density as a function of both time
and frequency. Consequently, flux density measurements made over a short time
and a narrow frequency range may differ substantially from the true mean flux
density. For this reason, most tabulated pulsar flux densities are subject to rather
large uncertainties, typically a factor of 2 or so.
1 0.3 CLUSTERS AND GALAXIES
10.31 Open Cluster Data
The data selection contained in the table "Selected Open Clusters" is a subset of G.
Lynga's Lund Catalog of Open Cluster Data (fifth edition, 1988). This computer-
based catalog is available from the World Data Center A or from Centre de Donnees
Stellaires. The fifth edition of the Lund catalog represents a new approach to the
presentation of available data. When several data values have been published in the
literature, the catalog contains weighted averages of these. The weights are derived
objectively using methods devised by Janes, Tilley, and Lynga, (1988). The data
collection for this work ended in September 1986.
The Lund catalog contains some 20,000 data values for a total of 1100 entries.
A separate file includes more than 500 references to authors and titles of primary
sources. The Astronomical Almanac table contains a selection from the Lund catalog
of relevant data values for the best studied clusters.
10.311 Definition of Open Clusters The clumping of stars in space is a physical
reality. Probably more than 50% of all stars are considered double or multiple stars.
Groups of ten or more stars are associations or clusters. Because the stars in each
such group share origin and conditions, they are of great use in the study of stellar
properties and in the study of properties of the larger systems of which they form
parts.
Associations are stellar groups that are less than 1 million years old. These
groups are often dynamically unbound systems that disperse into single stars, dou-
ble stars, and small groups. Occasionally, however, the associations are dynamically
10 / STARS AND STELLAR SYSTEMS 523
bound, in which case they survive to become open clusters (the ages of which range
from less than 1 million years to more than 1 billion years). Open clusters are quite
distinct from globular clusters, which are always dynamically bound and are several
billion years old. Open clusters form a flat disk subsystem, whereas the globular
clusters form an extended spheroid subsystem.
In the table of selected open clusters, there are a number of associations that
might not develop into physically bound clusters. The Trapezium cluster and some
of the Collinder objects belong to this category. Others, such as the Blanco and
Upgren clusters, are considered physical entities because their members have similar
motions and physical properties, although they are only a small part of the total
stellar population inside the volume they occupy. An effort has been made to exclude
from the catalog such specious clusters that are in fact only the apparent result of
the distribution of obscuring clouds.
10.312 Open Clusters as Related to Galactic Structure The system of open clus-
ters provides the best material for the study of the disk of our Galaxy. There
are several reasons for this. Distances, ages, and interstellar extinctions are better
known for open clusters than for other objects; open clusters have a range of ages
that make them particularly suited for study of the recent evolution of the galac-
tic system; and younger clusters congregate in spiral features. In fact, photometric
studies of open clusters are often carried out in the course of investigations of galac-
tic structure, and this also shows in the selection of objects. Thus, a lot of work has
been carried out on younger clusters and on very old open clusters, even though
some quite close-by open clusters of intermediate age are still relatively unknown.
10.3 13 Open Clusters as Related to Studies of Stellar Evolution Many of the break-
throughs in the study of stellar evolution have been aided by observations of open
clusters. The important point is that the cluster members are all of approximately
the same age and chemical composition. The only fundamental difference is in the
mass of gas from which each member is formed. The fact that more massive stars
get higher central temperature, higher luminosity, and a faster rate of evolution
than less-massive stars, has been utilized for most of the fundamental studies of
the principles of stellar evolution. Practical, observational advantages also result
from the fact that the cluster stars can be assumed to have the same distance from
us and, usually, show the same interstellar extinction.
10.314 Descriptional Data for Open Clusters An open cluster is defined by its po-
sition in the sky and by its apparent angular diameter. Each cluster has also tradi-
tionally been given one or several names referring to various catalogs or discoverers.
In fact, for some objects there is quite a profusion of different denotations. To help
alleviate confusion, the cluster commission of the International Astronomical Union
decided in 1979 on a nomenclature consisting of a "C" for cluster, four digits for
524 EXPLANATORY SUPPLEMENT
hours and minutes of right ascension (1950.0) and a plus or minus sign plus three
digits for degrees and tenths of degrees of declination (1950.0). In the Lund open
cluster catalog, the IAU designation and the most commonly used name are entered
for each cluster. The current, equatorial position is also entered.
The published values of angular diameters of clusters sometimes differ by a
large factor. This is partly due to different limiting magnitudes for different studies
but also to different opinions about the extent of each cluster. It is important for
comparative studies to have a consistent value for the diameter — in the fifth edition
of the Lund catalog this was accomplished using long exposure prints and films from
the existing series of Sky Surveys. The angular diameters are in minutes of arc and
refer to the cluster cores.
During the examination of the Sky Survey Charts, Trumpler classes were also
estimated. The classification scheme proposed by R.J. Trumpler (1930) contains
the following classes:
Concentration of cluster stars
I detached, strong central concentration
II detached, little central concentration
III detached, no noticeable concentration
IV not well detached, appearing like a star-field concentration
Range of stellar brightness
1 most stars of nearly the same brightness
2 medium range in brightness
3 bright and faint stars in the cluster
Richness of cluster
p poor
m moderately rich
r rich
Nebulosity
n cluster is involved in nebulosity
10.315 Derived Data for Open Clusters Data concerning distances, extinctions,
and physical properties of open clusters are determined during quite extensive in-
vestigations. It is fortunate that so much work is devoted to such tasks all over the
world. The work of the data compiler is to compare the published values, to analyze
their reliability, and to find the best available figure for each parameter.
The most reliable way to determine the distance of an open cluster is to observe
its color-magnitude diagram and fit the diagram to the main sequence of nearby
stars. When a cluster includes a number of unevolved stars on the main sequence,
this method of determining distances is in fact superior to other methods used for
10 / STARS AND STELLAR SYSTEMS 525
distant stars or star systems. For the distance of a nearby cluster with a long main
sequence, an imprecision as low as 10% can be achieved.
In connection with each photometric study, a determination of the interstellar
extinction in front of the cluster also has to be made. The usual method is to de-
termine the color excess due to the extinction and assume that the reddening ratio
between the extinction in visual light and the color excess is 3.1. A complication
is that the extinction varies over the cluster in some cases. Apart from those clus-
ters, and if the assumption about reddening ratio is correct, the extinction can be
determined to within about 0.1 magnitudes.
Frequently, the same investigations that determine the distances to clusters will
also provide estimates of their ages by using the turn-off point of the main sequence
from the zero-age main sequence. This is the most reliable method, and results
from it are given in The Astronomical Almanac table. To have a homogeneous set
of cluster ranges Janes, Tilley, and Lynga derived a calibration of the logarithm
of age as a function of the turn-off color. The listed values around 6.0— i.e., one
million years can, in most cases, be considered as rough upper limits. Clusters of
that age group might still be in the process of star formation and thus have a large
spread in age between member stars. Clusters with logarithmic ages higher than
7.0 have an imprecision of a few tenths of unit.
Data for the magnitude of the brightest cluster members and for the spectral
class of the earliest-type cluster members are fairly unambiguous. The only problem
lies in the possible inclusion of nonmembers. Naturally, efforts have been made to
avoid this source of error. The total magnitudes of the clusters have been calculated
by adding, logarithmically, the magnitudes of all known cluster members. This was
done and the data privately communicated to the author by B. Skiff of Lowell
Observatory. The open cluster table of The Astronomical Almanac also lists the
value of metallicity, which was determined by various methods in more than 30
different investigations and transferred to a consistent set of data by Janes, Tilley,
and Lynga. The definition of the metallicity ratio is the logarithmic ratio of the
abundances of iron and hydrogen divided by the corresponding ratio for the solar
spectrum. In practice, metal fines other than those of iron influence the parameter.
10.316 The Reliability of Open Cluster Data Despite great efforts to assess open
cluster data critically, it is quite clear that the set can never be quite homogeneous.
The variations in brightness of the objects in crowded fields and in observing con-
ditions contribute to an intrinsic variation in quality. In the Lund catalog, several
of the physical parameters have been assigned weights that describe in a compara-
tive fashion the precision of the data. The principles for weight assignment and the
analysis have been described in Janes, Tilley, and Lynga (1988).
526
1 0.32 Globular Star Cluster Data
EXPLANATORY SUPPLEMENT
The data on globular clusters given in The Astronomical Almanac are derived from
a variety of sources. The list of clusters itself is from the IAU's "Star Clusters
and Associations," which incorporated the official IAU cluster designations (the
four-digit, sign, three-digit numbers prefixed by a "C"), The original IAU list has
been supplemented by R.E. White as a result of his triennial literature searches to
prepare the "Work Published and In Progress" cluster-by-cluster summary for each
IAU Reports on Astronomy since 1975. These reports are not compiled after the
1988 IAU General Assembly. Instead, J.-C Mermilliod, at the Geneva Observatory,
continues to maintain a database of literature over open and globular clusters. A
copy resides at the National Optical Astronomical Observatories in Tucson, AZ.
The clusters' equatorial coordinates are taken from Shawl and White's (1986)
listing (preferred) or from the compilation by Webbink (1985). The axial-ratio data
(b I a)- values are taken exclusively from White and Shawl's (1987) extensive list.
10.321 The Definition of a Globular Star Cluster The celestial objects called glob-
ular star clusters cannot be identified by a casual inspection of a photographic plate,
instead certain criteria must be met. Foremost among the "physical-appearance"
criteria is a cluster's contrast against the local field: The globular star cluster should
be seen as a higher image-density object than the surrounding neighborhood of field
stars. The cluster should also appear to be somewhat compactly organized within its
own image. In addition, a gradient of decreasing areal star density with increasing
radius from the cluster center should be apparent.
Globular star clusters often contain short-period variable stars of the RR Lyrae
class; open star clusters and stellar associations, however, never contain such stars.
The brightest stars in globular clusters are red giants, and are seen spectroscopically
to be weak in their abundance of heavy elements when compared to the solar
abundance of same.
However, amongst the globular clusters resident in the Milky Way Galaxy,
exceptions exist to all of the aforementioned criteria. The only unambiguous de-
termination of "globular cluster-ness" is made using the cluster's color- magnitude
diagram (CMD) — globular clusters have CMD's that range in the values of (B-V)
color from, approximately, —0.6 to +1.6. There is, roughly, a 2 to 3 magnitude
difference between a globular cluster's main sequence turn-off and the apparent
magnitude level of its horizontal branch (HB). A similar magnitude range exists
between the tip of the red-giant branch and the HB, although this difference is
subject to variation as a function of a cluster's intrinsic metal abundance; the more
metal-abundant, the smaller the difference (but not less than 2 mag.).
Kinematically, the globulars form a two-component group, again as a function
of their metal abundances. The metal-rich group (between 10 and 50% of the solar
1 0 / STARS AND STELLAR SYSTEMS 527
value) are seen to be in nearly circular orbits that are confined to the galactic disk.
The metal-poor group (the preponderance of the 146 known globulars) appear to be
moving in highly eccentric elliptical paths around the center of the Galaxy, in orbits
with planes that may be highly inclined to the general plane of the disk (although
the orientation of the clusters' orbital planes to the disk seems to be quite random).
The clusters orbit the galactic center in a conventional right-handed sense, the only
exception (so far) being NGC 3201 (C1015-461), which appears to be in a retrograde
galactic orbit.
10.322 The Importance of Globular Clusters for Galactic Structure Historically,
the globular clusters were used by Harlow Shapley to determine the Sun's distance
from the center of our Galaxy. Shapley's work led to the realization that our Sun is
not in the Galaxy's center, as had been cheerfully assumed up to then, but is located
about two-thirds of the radial distance outward from the center. The circumstances
that led to Shapley's discovery, namely the presence of RR Lyrae variable stars
within many of the globulars, still makes these clusters important for large-scale
dynamical and kinematical studies of our Galaxy. Because many external galaxies
also show evidence of a globular star cluster population, the Galaxy's clusters are
also useful as secondary distance estimators to these more distant systems.
10.323 The Importance of Globular Clusters for Stellar Evolutionary Studies Be-
cause the many stars in a globular cluster are at the same distance from the Sun,
a globular cluster CMD involving apparent (visual) magnitude versus visual color-
index reveals the same pattern within that parametric space as the equivalent dis-
tribution of stellar characteristics within the luminosity versus surface-temperature
parametric space. Therefore, the globular clusters provide the theorist with test-bed
situations for comparing metal-weak model stars against low-metal observational
data. The globular clusters have been revealed as being some of the oldest objects
in the Galaxy, with ages ranging from 10 to 15 billion years. Hence, observational
studies of stellar associations provide comparative data for young, metal-rich stars;
and open clusters are the source of data for metal-rich stars intermediate in age.
Globular clusters complete the observational set by providing astrophysical param-
eters for the old and metal-poor stars formed in the initial evolutionary stages of
our Galaxy.
10.324 Basic References for Globular Clusters Studies There has not been a sum-
mary monograph written on the general subject of the globular stars clusters for
over 25 years (Arp 1965; Hogg 1959); however, because new information on clusters
528
EXPLANATORY SUPPLEMENT
is developing so rapidly, and often in such unexpected directions (e.g., X-ray emis-
sion from point sources within clusters), writing an all-encompassing monograph
would be futile at present. To come to grips with current states of research areas
concerning the clusters, one must go preferentially to the colloquia and symposia
held under the auspices of the IAU.
General summaries on the subject are given by Arp (1965) and Hogg (1959).
Data collections are found in Kukarkin (1974), Hogg (1973), Philip et al. (1976),
Shawl and White (1986), and White and Shawl (1987). Recent colloquia and sym-
posia are IAU Colloquium No. 68 (Philip and Hayes, 1981), IAU Symposium No,
113 (Goodman and Hut, 1985), NATO Advanced Study Institute, 1978 (Hanes and
Madore, 1980), IAU Symposium No. 126 (Grindlay and Philip, 1988), and IAU
Symposium No. 85 (Hesser, 1979).
10.33 Bright Galaxies
The 194 galaxies selected for inclusion in The Astronomical Almanac are generally
brighter than 5-band magnitude 11.5 and/or larger than 4.5 arcminutes. These
galaxies were extracted from an up-dated version of the Second Reference Catalogue
of Bright Galaxies (RC2) (de Vaucouleurs et al. 1976).
A galaxy's morphological type is based on the revised Hubble system described
by de Vaucouleurs (1959a, 1963). The system includes four classes: Ellipticals (E),
Lenticulars (L), Spirals (S), and Irregulars (I). In the L and S classes (disk galax-
ies), two families, ordinary (A) and barred (B), are distinguished, with transition
types denoted AB. Each family can exist in two varieties, depending on the absence
(s) or presence (r) of an inner ring structure, with transition types denoted (rs). In
the L class, the relative importance of the two major components of a galaxy — the
spheroid (or bulge) and the disk — is denoted by the superscript "— " (early, i.e.,
most similar to E), "0" (intermediate), and "+" (late, i.e., more similar to S). The
stage along the Hubble sequence of spirals is denoted by a, b, c, and d in order of
decreasing bulge-to-disk ratio and increasing development of the arms. The tran-
sition type between L and S is denoted SO/a; between S and magellanic irregulars
(Im), by Sm. Thus, the Large Magellanic Cloud is the prototype of SB(s)m, and
the Andromeda galaxy (M31) of SA(s)b. An outer ring structure is designated by
the symbol (R) preceding the Hubble-stage designation. The puzzling nonmagel-
lanic irregulars, of which M82 is the prototype, are designated 10. Edge-on systems
are signaled by the suffix "sp" (for "spindle"). Peculiarities are noted by the suffix
"pec." For more detailed discussions of galaxy morphology, see Sandage (1975) and
Buta (1989).
Because physical properties, such as color index, hydrogen index, bulge-to-disk
ratio, and so on, are closely correlated with stage along the Hubble sequence, it is
convenient to attach to the stage a numerical scale, T, from —5 at E, through 0
at SO/a, to 10 at Im. The luminosity class, L, introduced by van den Bergh (1960
10 / STARS AND STELLAR SYSTEMS 529
a, b)- or better, the luminosity index A = {T + L) I 10 (de Vaucouleurs 1979) — is
correlated with absolute magnitude; both T and L are listed in the table. The L
scale, as used in RC2, runs from 1 for a bright giant galaxy, to 9 for a faint dwarf
system. The precision of T and L values is on the order of one unit or better. For
details, see de Vaucouleurs (1977) and de Vaucouleurs et al. (1978).
The isophotal apparent major diameter of galaxies is given in the table for the
isophote level 25.0 B-magnitude per square arcsecond (5-m/ss). The level corre-
sponds roughly to the visible size of the galaxy image seen on the Palomar Ob-
servatory Sky Survey paper prints (blue light), and to the diameters listed in the
Uppsala General Catalog (Nilson, 1973) or northern galaxies. Note that isophotal
diameters are not metric diameters. The former depend on the surface brightness
(specific intensity) of the galaxy image. The latter do not depend on surface bright-
ness, and while they are better indications of the true size of galaxies, they are more
difficult to measure and are not yet available for many galaxies. For more details,
see the Introduction to RC2. Note that to avoid negative logarithms, D25 is given
in the table in units of 0.1 arcminute.
The ratio of the major and minor axes at the isotope level 25.0 5-m/ss (R25 =
D25 t d.25) is also listed in logarithmic form. The precision is usually better than 0.1.
The isophotal axis ratios given in the table are corrected for the systematic errors
that affect visual measurements of photographic plates (Holmberg 1946, de Vau-
couleurs 1959b). (For more details on systematic and accidental errors in diameter
and axis ratios, see Paturel et al., 1987.)
The total apparent S-band magnitudes, Bj, are generally derived from photo-
electric measurements or detailed photographic surface photometry in the Johnson
B system. These magnitudes are weighted means of the best available magnitudes
extrapolated to infinity to allow for the generally small fraction of the luminosity
contributed by the outermost regions of a galaxy as explained in the Introduction to
RC2. The precision is generally 0.1 mag or better. (For details, see de Vaucouleurs
and Bollinger, 1977.) For some applications, the observed magnitudes need to be
corrected for galactic and internal extinction, which are both controversial and un-
certain by 0.1 to 0.2 mag or more in extreme cases (e.g., an edge-on spiral galaxy
near the galactic plane).
The total color indices, (B - V)T and (U-B)Tl in the Johnson U, B, V systems
are as observed; i.e., uncorrected for galactic and internal extinction. The range is
roughly from; +1.0 to +0.3 for (B-V)T, and +0. 6 to -0. 3 for (U-B)T as one moves
along the Hubble sequence from E (yellow-orange) to Im (whitish) through L and
S types. By an oversimplification, the extreme types are often described as "red"
and "blue" galaxies. The colors give crude indications of the energy distribution
in the continuum and, hence, of the stellar composition of a galaxy. However, the
colors are often modified by line emission from Hll regions (which decreases the
color index, or increases the blueness) and internal dust extinction (which increases
the color index, or increases the redness).
530 EXPLANATORY SUPPLEMENT
The heliocentric radial velocity, derived from optical observations and 21-cm
neutral hydrogen radio emission, is given in kms-1 and calculated with the optical
convention that AA / A = V I c. For the small redshifts involved, this convention is
entirely adequate. For a discussion of systematic and accidental errors in radial ve-
locities, see the Introduction to RC2. The radial velocity, V0 — corrected for galactic
rotation according to the old 1AU convention, V0 = V + 300cosA, where A - the
angle to the solar apex^is also listed. For cosmological applications, it is better
to replace the conventional solar motion of 300 kms^1 toward / = 90°, b = 0° (see
Trans. IAU, XVIB, 201, 1977), with the total velocity vector of 366 kms-1 toward
/ = 265°, b = +55° relative to the cosmic background radiation (see de Vaucouleurs
et al. 1981). For more detailed discussions of solar motion relative to different ex-
tragalactic frames of reference, see de Vaucouleurs and Peters (1984).
More up-dated data on bright galaxies can be found in the Third Reference
Catalogue of Bright Galaxies (de Vaucouleurs et al. 1991).
10.34 Quasi-Stellar Objects
A quasi-stellar object (hereafter, QSO) can be defined as an object with a large
redshift whose optical morphology is "quasi-stellar," (or "starlike" in the sense
that it appears as an unresolved point of light in a typical optical image). Note
that the terms QSO and quasar are often used synonymously, but the latter refers
technically only to the subset of QSOs that are strong radio sources (quasar being
a contraction of quasi-stellar radio source).
QSO names are as diverse as the surveys by which they were discovered. Most
QSOs are found either by virtue of their radio emission or in optical surveys designed
to select objects with strong emission-lines and/or unusual colors. The QSOs are
designated by a number internal to a survey (e.g., 3c 279 is object 279 in the third
Cambridge radio survey and UM 148 is object 148 in the University of Michigan
optical survey) or by appending the approximate 1950.0 coordinates to the acronym
denoting a survey (e.g., PKS2203-18 is taken from the Parkes radio survey and
PG0026+12 from the Palomar-Green optical survey). Note that there is wide range
in the accuracy of the published coordinates for QSOs, but a significant fraction
(especially the radio-selected QSOs) have positions known to better than one arcsec.
The most critical parameter for a QSO is its redshift,
z = (A - A0) / A0, (10.34-1)
where A0 and A are laboratory wavelength and observed wavelength respectively
of a particular spectral feature. QSO redshifts are almost always measured using
the strong, broad emission-lines of hydrogen, helium, and other heavier, cosmically
abundant elements that are the trademark of a QSO. In most cases, several emission
10 / STARS AND STELLAR SYSTEMS
531
lines are used to determine the redshift, but in some cases only a single emission-
line is present (usually assumed to be the hydrogen Lya line at A0 = 1216 A, the
strongest emission line in a typical QSO spectrum). In such cases, the redshift is
evidently uncertain. Measured QSO redshifts range from « 0. 1 to > 4. Note that
nuclei of many galaxies at still lower redshifts contain objects that (apart from their
low redshifts and modest luminosities) are essentially identical to QSOs. For his-
torical reasons, these are usually called active galactic nuclei {or, more specifically,
Type 1 Seyfert Nuclei) and are not considered to be "true" QSOs.
In addition to the broad emission lines, many QSOs (particularly those with
large redshifts) have a host of narrow absorption lines. The most common lines are
due to hydrogen Lya, but the UV resonance lines of cosmically abundant heavy ions
(e.g., C+3, Mg+1) are also often detected. These fines almost always have redshifts
that are much smaller than the emission-line redshift. They are believed to be
produced in intervening galaxies or intergalactic gas clouds.
Although the spectral-energy distributions of some QSOs have been studied
in considerable detail, most have only very limited data available (typically UBV
magnitudes and colors). Radio-selected QSOs have radio flux densities at several
frequencies and a radio spectral index (a) also available. The radio flux densities
are given in Janskys (Jy) or milfi-Janskys (mJy), where 1 Jy - 10 26 Watts m
Hz^'. The spectral index is defined by S(v)av~% where S is the flux density at a
frequency v.
The most extensive and recent compilations of QSOs and their basic proper-
ties are Hewitt and Burbidge (1987) and Veron-Cetty and Veron (1989). A good
introduction to the basic phenomenology of QSOs is given by Weedman (1986).
1 0.4 SOURCES CATEGORIZED BY WAVELENGTH REGION
10.41 Radio-Source Positional Calibrators
The list of 233 radio-source positions, given in The Astronomical Almanac, is from
Argue et al. (1984). This list was compiled by a working group under IAU Commis-
sion 24 as a first step in defining a catalog of extragalactic objects that have both
radio and optical counterparts. Positions in the fist were compiled from a number ot
previously published catalogs. The origin of right ascension is defined by the right
ascension of 1226+023 (3C273B) at epoch J2000.0, 12h26m06!6997, as computed by
Kaplan et al. (1982), and based on the B1950.0 position determined for the source
by Hazard et al. (1971). An indication of a position's uncertainty is given by the
number of digits in the tabulated coordinates; the end figures may be subject to
revision. The column headed S5aHz gives the flux density in Janskys at 5 GHz.
Fluxes of many of the sources vary, however, and the tabulated flux is meant to
serve only as a rough guide.
532
10.42 Radio-Flux Calibrators
EXPLANATORY SUPPLEMENT
The calibration of radio-source flux densities, in absolute terms, is an essential part
of observational radio astronomy. The basic absolute calibration requires the use of
small antennas, the gain of which can be theoretically calculated or measured (see
Findlay, 1966). Such small antennas can be used only to determine the flux densities
of the strongest three radio sources (Cas A, Cyg A, and Tau A). These sources are
unsuitable for the calibration of large antennas, particularly at high frequencies,
since they all have finite sizes and are all partially resolved. It is therefore necessary
to have a list of secondary calibrators, the flux densities of which are determined
with reference to the strongest sources as primary standards. The flux densities
of the sources given in The Astronomical Almanac have been determined in this
manner.
The data in the table were taken from Baars et al. (1977) and derived in the
following manner:
(1) The absolute spectrum of Cas A, with its secular rate of decrease taken into
account, was derived for epoch 1980.0 between frequencies of 300 MHz and
30 GHz.
(2) The absolute spectra of Cyg A and Tau A were derived from direct mea-
surements and from measurements of the ratios of their flux density to that
of Cas A.
(3) An accurate spectrum for the somewhat weaker source Vir A was estab-
lished from direct, accurate ratios of its flux density to those of Cas A and
Cyg A.
(4) This Vir A spectrum was used as a basis for obtaining accurate relative
spectra of the sources in the table, these sources being suitable for routine
calibration of flux densities.
Only three sources are suitable for the calibration of interferometers and syn-
thesis telescopes — 3C48, 3C147 and 3C286 (the positions for these are taken from
Elsmore and Ryle, 1976). Some other sources may need a correction for angular
size when used with the largest single antennas at high frequencies, but since the
sizes are well-known, the correction is easy to apply.
It must be emphasized, first, that the given flux densities cover a range from
400 MHz to 22 GHz: extrapolation beyond this range is dangerous. Second, al-
though these sources have "normal" spectra, characteristic of transparent syn-
chrotron sources, some do contain compact, opaque components that may be vari-
able. For example, 3C48 appears to show some variability at a level between 1
percent around 1.6 GHz and 30 percent around 20 GHz (Perley and Crane, 1985).
3C147 shows variations in its structure on milliarcsecond scales (Preuss et al., 1984)
and its flux density is also variable (Simon et al, 1983, Andrew et al., 1981). Accu-
rate calibrations of flux density measurements should make use of several sources
from the table, so that such effects may be detected and allowed for.
10 / STARS AND STELLAR SYSTEMS
10.43 X-Ray Sources
533
The X-ray sources given in The Astronomical Almanac were selected from those
having optical, radio, or infrared counterparts listed in an unpublished Catalogue
of X-Ray Positions (Dolan, 1983). The master list of sources in this survey file
is available on magnetic tape from the Astronomical Data Center, Mail Code 633,
NASA Goddard Space Flight Center, Greenbelt, MD, 20771. Access to the cat-
alog is also available through electronic mail via Bitnet at W3WHW@SCFMVS
or via SPAN at NSSDCA::WARREN. The original references for the position and
identification of each source are contained in the notes to the catalog.
Three general surveys of X-ray sources are the 4U catalog (Forman et al., 1978)
the 2A catalog (Cooke et al., 1978), and the 1H catalog (Wood et al., 1984). The
sources in these catalogs are designated by their X-ray location in the equatorial
coordinate system (equator and equinox 1950.0). The locations are given as right
ascension in hours and minutes and declination in degrees or degrees and tenths, and
are prefixed by the satellite initial letter and the edition number of the catalog (4U
= Fourth Uhuru catalog; 2A = Second Ariel catalog; 1H = First HEAO-1 catalog).
Hence, 4U1656+35, 2A1655+353, and 1H1656+354 all designate the same source
in the three different catalogs.
The nomenclature of X-ray source designations is confusing, primarily for his-
torical reasons. The brightest sources in the X-ray region are designated by constel-
lation in order of discovery. The letters XR (Friedman et al., 1967) or X (Giacconi
et al., 1967) are appended and followed by a number. Thus, Cyg XR-1 and Cyg
X-l are the same source. Certain sources in the Large and Small Magellanic Clouds
receive the same designation with LMC or SMC instead of a constellation as a pre-
fix. Another common usage designates sources by galactic longitude and latitude
prefixed by the letters GX (Bradt et al., 1968). Still other sources are designated
by prefixing the initials of the discover's satellite, institution, or name to a running
catalog number or a 1950.0 equatorial position. IAU Commission 48 decided m
1976 that no scheme of nomenclature was preferred over any other when referring
to sources designated in more than one system. The discovery designation is used in
The Astronomical Almanac. When no discovery designation is listed, the source is
always referred to by the common name of the identified counterpart. The IAU rec-
ommended in 1983 that sources newly discovered after that date be designated by
their X-ray position in right ascension and declination in the equatorial coordinate
system of 1950.0.
Tabulated positions are based on published positions of identified counterparts.
The (2-6) keV flux, in units of 10"11 erg cm"^-' (10-14 watt m'2), is taken from
the 4U catalog. For sources with variable X-ray intensities, the maximum observed
flux from the 4U catalog is tabulated. The tabulated magnitude is the optical
magnitude of the counterpart in the V filter, unless marked by an asterisk, in
which case the B magnitude is given. Variable magnitude objects are denoted by
534 EXPLANATORY SUPPLEMENT
V — for these objects the tabulated magnitude pertains to maximum brightness.
Codes specifying the class of objects to which the identified counterpart belongs
are explained at the end of the table.
10.5 REFERENCES
Andrew, B.H., MacLeod, J.M., and Feldman, PA. (1981). "Standard Sources at 10.6GHz
and Variability in 3C147" Astron. Astrophys. 99, 36-38.
Argue, A.N., de Vegt, O, Elsmore, B., Fanselow, J., Harrington, R., Hemenway, P., John-
ston, K.J., Kuhr, H., Kumkova, L, Niell, A.E., Walter, H., and Witzel, A. (1984). "A
Catalog of Selected Compact Radio Sources for the Construction of an Extragalactic
Radio/Opticai Reference Frame" Astron. Astrophys. 130, 191-199.
Arp, H.C. (1965). "Globular Clusters in the Galaxy" in Galactic Structure A. Blaauw
and M. Schmidt, eds. Vol. V of Stars and Stellar Systems G.P. Kuiper and B.M.
Middlehurst, eds. (University of Chicago Press: Chicago), pp. 401-434.
Baars, J.W.M., Genzel, R., Pautiny-Toth, I.I.K., and Witzel, A. (1977). "The Absolute
Spectrum of Cas A; An Accurate Flux Density Scale and a Set of Secondary Calibra-
tors" Astron. Astrophys. 61, 99-106.
Batten, A.H. (1985). "Radial- Velocity Standards" in Stellar Radial Velocities, IAU Coll.
No. 88 A.G.D. Philip and D.W. Latham, eds. (L. Davis Press: Schenectady).
Batten, A.H., Fletcher, J.M., and MacCarthy, D.G. (1989). "Eighth Catalogue of the
Orbital Elements of Spectroscopic Binary Systems" Pub. Dominion Astrophysical
Observatory XVII, 1-317.
Bessell, M.S. (1986). " VRI Photometry III: Photographic and CCD R and / Bands and
the Kron-Cousins RI System" Pub. Astron. Soc. Pac. 98, 1303-1311.
Bessell, M.S. and Weis, E.W. (1987). "The Cousins and Kron VRI Systems" Pub. Astron.
Soc. Pac. 99, 642-644.
Bidelman, W.P. (1969). "Stellar Spectra and Spectral Types" in Stellar Astronomy, H.-Y.
Chiu, R.L. Warasila, and J.L. Remo, eds. (Gordon and Breach, New York), Volume
1, pp. 147-206.
Bidelman, W.P. (1991). Private communication.
Bradt, H., Naranan, S., Rappaport, S., and Spada, G. (1968). "Celestial Positions of X-Ray
Sources in Sagittarius" Astrophys. J. 152, 1005-1013.
Buta.R.J. (1989). "Galaxy Morphology and Classification" in The World of Galaxies H.G.
Corwin and L. Bottinelli, eds. (Springer Verlag, New York), pp. 29-47.
Campbell, B., Walker, G.A.H., and Yang, S. (1988). "A Search for Substellar Companions
to Solar-Type Stars" Astrophys. J. 331, 902-921.
Cooke, B.A., Ricketts, M.J., Maccacaro, T., Pye, J. P., Elvis, M., Watson, M.G., Griffiths,
R.E., Pounds, K.A., McHardy, I., Maccagni, D., Seward, F.D., Page, C.G., and Turner,
M.J.L. (1978). "The Ariel V (SSI) Catalog of High Galactic Latitude (|b| >10 deg.),
X-Ray Sources" Mon. Not. R. Astron. Soc. 182, 489-515.
Cousins, A.W.J. (1976). " VRI Standards in the E Regions" Mem. R. Astron. Soc. 81,
25-36.
Crawford, D.L. and Barnes, J.V. (1970). "Standard Stars for uvby Photometry" Astron.
J. 75, 978-998.
Crawford, D.L. and Mander, J. (1966). "Standard Stars for Photoelectric H-Beta Pho-
tometry" Astron. J. 71, 114-118.
10 / STARS AND STELLAR SYSTEMS
535
Curtis, R.H. (1932). "Classification and Description of Stellar Spectra" in Handbuch der
Astrophysik Part V (J. Springer, Berlin), pp. 1-108.
de Vaucouleurs, G. (1959a). "Classification and Morphology of External Galaxies" in
Handbuch der Physik S. Flugge, ed. (Springer- Verlag, Berlin), 53, 275-310.
de Vaucouleurs, G. (1959b). "Photographic Dimensions of the Brighter Galaxies" Astron.
J. 64, 397-409.
de Vaucouleurs, G. (1963). "Revised Classification of 1500 Bright Galaxies" Astrophys. J.
Suppl. 8, 31-97.
de Vaucouleurs, G. (1977). "Qualitative and Quantitative Classifications of Galaxies" in
The Evolution of Galaxies and Stellar Populations B.M. Tinsley and R.B. Larson,
eds. (Yale Univ. Observatory, New Haven), pp. 43-96.
de Vaucouleurs, G. (1979). "The Extragalactic Distance Scale. V. Tertiary Distance Indi-
cators" Astrophys. J. 227, 380-390.
de Vaucouleurs, G., de Vaucouleurs, A., and Corwin, H.G., Jr. (1976). Second Reference
Catalogue of Bright Galaxies, (RC2), (University of Texas Press, Austin), pp. 1-396.
de Vaucouleurs, G. and Bollinger, G. (1977). "Contributions to Galaxy Photometry. VI.
Revised Standard Total Magnitudes and Colors of 228 Multiply Observed Galaxies"
Astrophys. J. Suppl. 34, 469-477.
de Vaucouleurs, G., de Vaucouleurs, A., and Corwin, H.G.,Jr. (1978). "Systematic and
Accidental Errors in Galaxy Luminosity Classification" Astron. J. 83, 1356-1359.
de Vaucouleurs, G., Peters, W.L., Bottinelli, L., Gouguenheim, L., and Paturel, G. (1981).
"Hubble Ratio and Solar Motion from 300 Spirals Having Distances Derived from H
I Line Widths" Astrophys J. 248, 408-422.
de Vaucouleurs, G., and Peters, W.L. (1984). "The Dependence on Distance and Redshift
of the Velocity Vectors of the Sun, the Galaxy, and the Local Group with Respect to
Different Extragalactic Frames of Reference" Astrophys. J. 287, 1-16.
de Vaucouleurs, G., de Vaucouleurs, A., Corwin, H.G., Buta, R., Paturel, G., and Fouque,
P. (1991) Third Reference Catalogue of Bright Galaxies, (Springer- Verlag, New York).
Dolan, F. (1983). Private communication.
Egret, D. and Wegner, M. (1988). "SIMBAD-Present Status and Future" in Astronomy
From Large DatabasesESO Conference and Workshop Proceedings No. 28 F. Murtagh
and A. Heck, eds. (European Southern Observatory, Munich), pp. 323-328.
Elsmore, B. and Ryle, M. (1976). "Further Astrometric Observations with the 5-km. Radio
Telescope" Mon. Not. R. Astron. Soc. 174, 411-423.
Explanatory Supplement to the Ephemeris (1961). Prepared jointly by the Nautical Alma-
nac Offices of the United Kingdom and the United States of America (Her Majesty's
Stationary Office, London).
Findlay, J.W. (1966). "Absolute Intensity Calibrations in Radio Astronomy" Ann. Rev.
Astron. Astrophys. 4, 77-94
Forman, W., Jones, C, Cominsky, L., Julien, P., Murray, S., Peters, G., Tananbaum, H.,
and Giacconi, R. (1978). "The Fourth Uhuru Catalog of X-Ray Sources" Astrophys.
J. Supp. 38, 357-412.
Friedman, H., Bryan, E.T., and Chubb, T.A. (1967). "Distribution and Variability of
Cosmic X-Ray Sources" Science 156, 374-378.
Garrison, R.F., ed. (1984). The MK Process and Stellar Classification (David Dunlap Obs.,
Toronto), pp. 1-423.
Giacconi, R., Gorenstein, P., Gursky, H., and Waters, J.R. (1967). "An X-Ray Survey of
the Cygnus Region" Astrophys. J. 148, L119-L127.
536
EXPLANATORY SUPPLEMENT
Goodman, J. and Hut, P., eds. (1985). Dynamics of Star Clusters, IAU Symposium No.
113 (Reidel, Dordrecht, Holland), pp. 1-622.
Gray, R.O. (1989). "The Extension of the MK Spectral Classification System to the In-
termediate Population II F-Type Stars" Astron. J. 98, 1049-1062.
Grindlay, J.E. and Philip, A.G.D., eds. (1988). Globular Clusters Systems in Galaxies, IAU
Symposium No. 126, The Harlow Shapley Centenary Commemorative Symposium
(Kluwer, Dordrecht, Holland), pp. 1-751.
Hanes, D. and Madore, B., eds. (1980). Globular Clusters Proceedings of NATO Advanced
Study Institute, Institute of Astronomy, August, 1978 (Cambridge University Press,
Cambridge, MA), pp. 1-390.
Hazard, C, Sutton, J., Argue, A.N., Kenworthy, CM., Morrison, L.V., and Murray, C.A.
(1971). "Accurate Radio and Optical Positions of 3C 273B" Nature Phys. Sci. 233,
89-91.
Hesser, J.E., ed. (1979). Star Clusters, IAU Symposium No. 85, (Reidel, Dordrecht, Hol-
land), pp. 1-516.
Hewitt, A. and Burbidge, G. (1987). "A New Optical Catalog of Quasi-Stellar Objects"
Astrophys. J. Supp. 63, 1-246.
Hewitt, A. and Burbidge, G. (1989). "The First Addition to the New Optical Catalog of
Quasi-Stellar Objects" Astrophys. J. Supp. 69, 1-63.
Hirshfeld, A. and Sinnott, R.W., eds. (1982). Sky Catalogue 2000.0, Volume 1, Stars to
Magnitude 8.0, (Sky Publishing, Cambridge, MA and Cambridge University Press,
Cambridge, England), pp. 1-604.
Hoffleit, D. and Jaschek, C. (1982). The Bright Star Catalogue, Fourth Revised Edition
(Yale University Observatory, New Haven), pp. 1-472.
Hogg, H.S. (1973). A Third Catalogue of Variable Stars in Globular Clusters Comprising
2119 Entries Volume 3, Number 6 (David Dunlap Obs., Toronto), pp. 1-75.
Hogg, H.S. (1959). "Star Clusters" in Handbuch der Physik, 53 S. Flugge, ed. (Springer-
Verlag, Berlin), pp. 129-207.
Holmberg, E. (1946). "The Apparent Diameters and the Orientation in Space of Extra-
galactic Nebulae" Medd Lund Astron. Obs. Series II, 117, pp. 1-82.
International Astronomical Union (1957). Transactions, Volume IX (Cambridge University
Press, Cambridge, England), p. 442.
International Astronomical Union (1973). Transactions, Volume XVA (Reidel, Dordrecht,
Holland), p. 409.
International Astronomical Union (1988). Transactions, Volume XXA (Reidel, Dordrecht,
Holland), p. 362.
International Astronomical Union (1990). Transactions, Volume XXB (Reidel, Dordrecht,
Holland), p. 267.
Irwin, A.W., Campbell, B., Morbrey, C.L., Walker, G.A.H., and Yang, S. (1989). "Long-
Period Radial- Velocity Variations of Arcturus" Pub. Astron. Soc. Pac. 101, 147-159.
Janes, K.A., Tilley, C, and Lynga, G. (1988). "Properties of the Open Cluster System"
Astron. J. 95, 771-784.
Johnson, H.L. (1963). "Photometric Systems" in Basic Astronomical Data K.A. Strand,
ed. Volume III of Stars and Stellar Systems, G.P. Kuiper and B.M. Middlehurst, eds.
(University of Chicago Press, Chicago), pp. 204-224.
Johnson, H.L., Mitchell, R.I., Iriarte, B., and Wisniewski, W.Z. (1966). "UBVRIJKL Pho-
tometry of the Bright Stars" Communications of the Lunar and Planetary Laboratory
4, 63, 99-241.
10 / STARS AND STELLAR SYSTEMS
537
Johnson, H.L. and Morgan, W.W. (1953). "Fundamental Stellar Photometry for Standards
of Spectral Type on the Revised System of the Yerkes Spectral Atlas" Astrophys. J.
117, 313-352,
Kaplan, G.H., Josties, F.J., Angerhofer, P.E., Johnston, K.J., and Spencer, J.H. (1982).
"Precise Radio Source Positions from Interferometric Observations" Astron. J. 87
570-576.
Keenan, PC. and McNeil, R.C. (1976). An Atlas of Spectra of the Cooler Stars (Ohio
State University Press, Columbus).
Keenan, P.C. and Morgan, W.W. (1951). "Classification of Stellar Spectra" in Astrophysics
J.A. Hynek, ed. (McGraw-Hill, New York), p. 12-28.
Kholopov, P.N., ed. (1985). General Catalogue of Variable Stars, fourth ed., Volume I
1985, pp. 1-376; Volume II, 1985, pp. 1-360; Volume III, 1987, pp. 1-368 (Nauka
Publ. House, Moscow).
Kron, G.E., Gascoigne, S.C.B., and White, H.S. (1957). "Red and Infrared Magnitudes
for 282 Stars with Known Trigonometric Parallaxes" Astron. J. 62, 205-220.
Kukarkin, B.V. (1974). General Catalogue of The Globular Star Clusters of a Galaxy
containing Physical Characteristics of 129 Objects Reduced to One System (Nauka
Press, Moscow), pp. 4-136.
Landolt, A.U. (1973). " UBV Photoelectric Sequences in the Celestial Equatorial Selected
Areas 92-115" Astron. J. 78, 959-981.
Landolt, A.U. (1983). " UB VR1 Photometric Standard Stars Around the Celestial Equa-
tor" Astron. J. 88, 439-460.
Lyne, A., Manchester, R.N., and Taylor, J.H. (1993). In preparation.
Lynga, G. (1988). "The Lund Catalog of Open Cluster Data" in Astronomy From Large
Databases, ESO Conference and Workshop Proceedings No. 28 F. Murtagh and A.
Heck, eds. (European Southern Observatory, Munich) pp. 379-382.
Manchester, R.N. and Taylor, J.H. (1977). Pulsars, (W.H, Freeman, San Fransisco). pp.
1-281.
Manchester, R.N. and Taylor, J.H. (1981) "Observed and Derived Parameters for 330
Pulsars" Astron. J. 86, 1953-1973.
McCarthy, M.F., Philip, A.G.D., and Coyne, G.V., eds. (1979). Spectral Classification of
the Future IAU Colloquium No. 47 (Vatican Obs, Vatican City), pp. 1-375.
Mermilliod, J.C. (1984). "Compilation of Photometric Data: Present State and Projects"
Bull. Inform. CDS 26, 3-7.
Morgan, W.W. and Keenan, P.C. (1973). "Spectral Classification" Annual Rev. of Astron-
omy and Astrophysics 11, 29-50.
Morgan, W.W., Abt, H.A., and Tapscott, J.W. (1978). Revised MK Spectral Atlas for
Stars Earlier than the Sun (Yerkes Observatory. Chicago and Kitt Peak Nat'l. Obs..
Tucson).
Morgan, W.W., Keenan, P.C., and Kellman, E. (1943). .4n Atlas of Stellar Spectra (Uni-
versity of Chicago Press, Chicago).
Nicolet, B. (1978). "Catalogue of Homogeneous Data in the UBV Photoelectric Photo-
metric System" Astron. Astrophys. Supp. 34, 1-49.
Nilson, P. (1973). Uppsala General Catalogue of Galaxies (Roy. Soc. Sci. Uppsala. Upp-
sala), pp. 1-456.
Ochsenbein, P. and Dubois, P. (1986). "Some Aspects of the Bibliography of Astronomical
Objects" Bull. Inform. CDS 31, 137-140.
538
EXPLANATORY SUPPLEMENT
Paturel, G., Fouque, P., Lauberts, A., Valentijn, E.A., Corwin, H.G., and de Yaucouleurs,
G. (1987). "Standard Photometric Diameters of Galaxies III. Reduction of the Diam-
eters in the ESO-B and SGC Catalogues to the Standard Diameter System at the 25
mag arcsec~2 Brightness Level" Astron, Astrophys. 184, 86-92.
Pedoussaut, A., Carquillat, J.M., Ginestet, N., and Vigneau, J. (1988). "Binaires Spectro-
scopiques. 15e Catalogue Complementaire" Astron. Astrophys. Suppl. 75, 441-443.
Perley, R. and Crane, P. (1985). "3C 48 Variability" NRAO Newsletter no. 25, p. 1.
Perry, C.L., Olsen, E.H., and Crawford, D.L. (1987). "A Catalog of Bright uvby(beta)
Standard Stars" Pub. Astron. Soc. Pac. 99, 1184-1200.
Philip, A.G.D., Cullen, M.F., and White, R.E. (1976). "IW Color- Magnitude Diagrams
of Galactic Globular Clusters" Dudley Observatory Reports (Dudley Observatory, Al-
bany), pp. 1-186.
Philip, A.G.D. and Hayes, D.S., eds. (1981). Astrophysical Parameters for Globular Clus-
ters, I.A.U. Colloquium No. 68 (L. Davis Press, Schenectady), pp. 1-614.
Preuss, E., Alef, W., Whyborn, N., Wilkinson, P., and Kellermann, K. (1984). "The Mil-
liarcsecond Core of 3C 147 at 6 cm" in VLBI and Compact Radio Sources, R. Fanti,
K, Kellermann, and G. Setti, eds. IAU Symp. No. 110 (Reidel, Dordrecht, Holland),
p. 29.
Rawley, L.N., Taylor, J.H., and Davis, M.M. (1988). "Fundamental Astrometry and Mil-
lisecond Pulsars" Astrophs. J. 326, 947-953.
Sandage, A. (1975). "Classification and Stellar Content of Galaxies Obtained from Direct
Photography" in Galaxies and the Universe A. Sandage, M. Sandage, and J. Kristian,
eds. Volume IX of Stars and Stellar Systems (University of Chicago Press, Chicago),
pp. 1-35.
Schmidt-Kaler, Th. (1982), "Calibration of the MK System" in Landolt-Bornstein, New
Ser., Gr. VI, 2b, pp. 14-24 and 451-456. ( Springer- Verlag, Berlin).
Shawl, S.J. and White, R.E. (1986). "Accurate Optical Positions for the Centers of Galactic
Globular Clusters" Astron. J. 91, 312-316.
Simon, R.S., Readhead, A.C.S., Moffet, AT., Wilkinson, P.N., Allen, B., and Burke B.F.
(1983). "Low Frequency Variability and Predicted Superluminal Motion in 3C 147"
Nature 302, 487-490.
Stromgren, B. (1963). "Quantitative Classification Methods" in Basic Astronomical Data
K.A. Strand, ed. Vol III of Stars and Stellar Systems G.P. Kuiper and B.M. Middle-
hurst, eds. (University of Chicago Press, Chicago), pp. 123-191.
Taylor, B.J. (1986). "Transformation Equations and Other Aids for VRI Photometry"
Astrophys. J. Supp. 60, 577-599.
Trumpler, R.J. (1930). "Preliminary Results on the Distances, Dimensions and Space
Distribution of Open Star Clusters" Lick Observatory Bulletin 14, 154-188.
van den Bergh, S. (1960a). "A Preliminary Luminosity Classification of Late-Type Galax-
ies" Astrophys. J. 131, 215-223.
van den Bergh, S. (1960b). "A Preliminary Luminosity Classification for Galaxies of Type
Sb" Astrophys. J. 131, 558-573.
Veron-Cetty, M.-P. and Veron, P. (1989). "A Catalog of Quasars and Active Nuclei" 4th
ed. ESO Scientific Report No. 7 (European Southern Observatory, Munich), pp. 1-143.
Webbink, R.F. (1985). "Structure Parameters of Galactic Globular Clusters" in Dynamics
of Star Clusters, J. Goodman and P. Hut, eds. IAU Symposium No. 113, (Reidel,
Dordrecht, Holland), pp. 541-577.
Weedman, D.W. (1986). Quasar Astronomy (Cambridge University Press, Cambridge,
England), pp. 1-217.
10 / STARS AND STELLAR SYSTEMS
539
Wegner, M., Egret, D., and Ochsenbein, F. (1989). SI MB AD User's Guide, (English ver-
sion by Wayne H. Warren, Jr., of NSSDC, and Joyce M. Watson, of SAO). (CDS,
Strasbourg)
White, R.E. and Shawl, S.J. (1987). "Axial Ratios and Orientations for 100 Galactic
Globular Star Clusters" Astrophys. J. 317, 246-263.
Wood, F.B., Oliver, J.P., Florkowski, D.R., and Koch, R.H. (1980). "A Finding List for
Observers of Interacting Binary Stars" Pub. University of Pennsylvania, Astron. Ser.
XII, pp. 1-349.
Wood, K.S., Meekins, J.F., Yentis, D.J., Smathers, H.W., McNutt, D.P., Bleach, R.D.,
Byram, E.T., Chubb, T.A., and Friedman, H. (1984) "The HEAO A-l X-Ray Source
Catalog" Astrophys. J. Supp. 56, 507-649.
CHAPTER
Computational Techniques
by R.L. Duncombe
1 1 .1 INTRODUCTION TO COMPUTING TECHNIQUES
In this section we will examine several computing techniques of special relevance to
astronomical calculation. For a more general treatment, see textbooks on computing
methods and numerical analysis. A short note on elementary computing principles
is also included in Interpolation and Allied Tables, which covers the sources of mis-
takes, the nature of checks, and the nature and effect of the inevitable errors due to
rounding-off and other causes. These topics, although of fundamental importance,
are not discussed here.
An important principle of computation is that the data used determines the
maximum precision of a calculation. This precision can be reduced by poor comput-
ing— for example, by a poor choice of formula or by the failure to retain an adequate
number of figures in the intermediate stages — but it can never be increased. Gen-
erally no simple relation exists between the errors (absolute or relative) of the final
result and of the data, though a relation can readily be seen numerically by follow-
ing through the calculation step by step, and often geometrically. It is misleading
to quote a result to more figures than is justified by the data on which it is based,
and it is erroneous to do so if an inadequate number of figures have been retained
in the intermediate stages.
Retaining more figures than necessary is a waste of effort if it significantly
increases the work involved. In some operations, however, especially in those done
on a computer or a personal computer, extra figures involve very little additional
work; but in other operations, an extra figure may require a more elaborate formula
or extra word length, resulting in much more additional work. A suitable number of
figures is that which offers the greatest convenience consistent with the building-up
error, due to accumulation of rouiuling-offs, not exceeding the error of the data;
this must be judged in relation to each calculation.
541
542
EXPLANATORY SUPPLEMENT
Figure 11.1.1
Spherical triangle
Mathematical formulas, if correct, adequately give a desired precision in the
result; but they may be inefficient, inconvenient, and misleading. One example is
the inefficiency of determining a small angle from its cosine; for a stated precision of
angular measurement the number of decimals in the cosine is inversely proportional
to the angle; and for a fixed number of decimals in the cosine (the usual and
convenient case) the precision obtainable decreases as the angle decreases. Consider,
for example, the formula for the third side of a spherical triangle, (Figure 11.1.1)
given two sides and the included angle,
cos a = cos b cos c + sin b sin c cos A = z, (11.1-1)
in which a is to be determined from measured values of b, c, and A. When a is
small, it can be found to the same precision as the data only by the apparently
incorrect procedure of retaining more decimals in each trigonometric function on
the right-hand side (z) of the equation than the data appear to justify. This is
legitimate because both b-c and A must be small so that their cosines are known
to extra figures. There are circumstances, however, in which the inconvenience of
such a procedure is outweighted by other factors.
The apparent failure of a mathematical formula should not be confused with
real geometrical limitations of precision. In the spherical triangle ABC,
sin a sin B = sin b sin A = x,
sin a cos Z? = cos 6 sin c — sin cos c cos A = y\
(11.1-2)
11 / COMPUTATIONAL TECHNIQUES
543
bot h x and y are small (y by the cancellation of two nearly equal components) if a is
small. It is not legitimate here to use extra figures in the trigonometric functions in x
and y, and B can be found with a precision only proportional to cosec a; this precision
is, however, clearly adequate to fix the point C (from BA). Similar arguments hold
for the angle C. Neither B nor C can individually be determined as precisely as 6,
C, or A; but the sum B + C can be so determined, as may be seen from geometric
considerations and from the equation
1 cos Ub + c) 1
cot-(fl + C)= f- tan -A. (11.1-3)
2 cos ~{b -c) 2
Important problems in astronomy may require the sum (or difference) of two angles
to be known more precisely than is possible with either (for example, the elements
of a planetary orbit); therefore, great care is required in handling them. The rec-
ommended method is to determine one of the angles and then treat it as exact in
finding the other, which must not be found independently.
Although there may be some uncertainty of precision in determining an angle,
there should be no uncertainty of quadrant. This can always be achieved by ade-
quate choice of formulas. For example, the following formulas for the solution of a
spherical triangle, in which two sides and the included angle are given, are always
adequate if the sides are less than 180°.
sin a sin B = sin 6 sin A —x,
sin a cos B = cos b sin c — sin b cos c cos A = y,
cosa = cos£>cosc + sinfcsinccosA -z, (11.1-4)
sin a cos C = sin b cos c — cos b sin c cos A,
sin a sin C = sin c sin A.
The check x2 +y2 + r - 1 = 0 is useful. The sign of z determines the quadrant
of a, and the signs of x and y that of B\ similarly with C.
In astronomical problems, the sides and angles of triangles on the celestial
sphere may be of any magnitude, and it is undesirable to solve them by methods
that restrict the sides to arcs less than 180°. The preceding general formulas are
all valid for triangles with sides of any length, and may be applied immediately to
the general triangles of spherical astronomy without any restriction to values less
than 180°. Furthermore, to make the solution determinate, it is only necessary to
find the algebraic sign of both the sine and the cosine of each arc or angle that may
exceed 180° to fix the quadrants in which they lie. Any of the cases of the general
triangle is determinate when, in addition to the three given parts, the algebraic sign
of the cosine of one of the required parts is also given. In most practical problems
544
EXPLANATORY SUPPLEMENT
Table 11.1.1
Precision of Angle and Number of Decimals for Trigonometric
Functions
1' =0.00029 1s =0.00007 0.0001 =0134 = 1f4
1" =0.0000048 0f1 = 7x10"6 1x10-6 = 0:'2 = Of 01 4
0:'1 = 5x10"7 0f01 =7x10-7 1x10-7=0:'02 =0f0014
0:'01=5x10-8 0f001=7x10-8 1 x 10-8 = 0:'002 = 0?00014
the conditions of the problem supply this sign. In general triangles the utmost care
should be taken to specify unambiguously the direction of measurement of angles
and arcs.
Collectively, the preceding formulas are sufficient for the solution of the general
triangle without restriction on the magnitudes of the parts, but, in practice, the
additional formulas
cos A - — cos B cos C + sin B sin C cos a,
sin A cos b = cos B sin C + sin B cos C cos a,
sin A cos c = sin B cos C + cos B sin C cos a,
sin/l cotB = sinccot^ - cosccosA,
sinAcotC = sin^cotc — cosfccosA, (11.1—5)
are very useful, (Collected formulas are given in Table 11.3.1 p. 550.)
In practice, the trigonometric functions of b, c, and A are taken out with a
number of decimals depending on the precision of the data, and the capacity of the
c alculating machine to be used. Table 11.1.1 gives corresponding precisions of angle
and number of decimals.
If no additional work is involved, it is clearly advantageous to use more decimals
than the precision warrants, since the effect of rounding-off errors is then much
reduced. For instance, seven decimals give just sufficient coverage for data known
to about 0'.'02 and for results to be rounded off to 0'T; but interpolation in eight-
figure tables at interval 1 " offers little, if any, more difficulty than interpolation in
seven- figure tables at interval 10", and eight, decimals can be used with little extra
work.
In the solution of Equations 11.1-4 and 11.1-5, a is found either from cosa = z
or from sina = (x2+y2)112, whichever is the smaller. Assuming a is less than 180°, the
ambiguity of the second formula is resolved according to the sign of z. B is found
from
tanB=- or cotfi=-, (11.1-6)
y x
11 / COMPUTATIONAL TECHNIQUES
545
Table 11.1.2
The Method of Inverse Use
0°
45°
90°
135°
180°
225°
270°
315°
360°
X
+
+
+
+
y
+
+
+
+
+
tan or cot
tan
cot
cot
tan
tan
cot
cot
tan
cot
D or C arg
D
C
D
C
D
C
D
C
D
according to which ratio is less than unity. The quadrant is determined by the signs
of x and y, since, if a is less than 180°, sina is positive.
Most trigonometric tables are arranged semiquadrantally, with direct (D) and
complementary (C) arguments. The Table 11.1.2 shows the method of inverse use
in finding B. There is no difficulty in systematic computation.
The range of precision of the inverse determination of an angle from its sine (or
cosine) and tangent (or cotangent) is indicated in Table 11.1.3. The table is arranged
to show the range of angle for which the alternative trigonometric function should
be used, and gives the error in the angle corresponding to an error of 1 x 10-6 in
the function.
The technique of inverse interpolation, referred to briefly in Section 11.23, is a
powerful tool in the solution of transcendental equations and of equations in which
the algebraic solution is complicated. The fundamental principle of such methods
lies in the tabulation of a discriminant, defined so that it attains a predetermined
value (usually zero) when the original equation is satisfied, and the calculation of
the unknown argument corresponding to the predetermined value of the discrimi-
nant by the process of inverse interpolation. Although these methods may involve
more calculation than direct methods, they have two considerable advantages. First,
they are usually independent of theoretical developments, of any approximations
that may be necessary in such developments, and of extrapolations. Second, the
Table 11.1.3
Range of Precision of the Inverse Determination of an Angle
Errors corresponding to
0°
45°
90°
135°
180°
1 x 10 6 in function
180°
225°
270°
315°
360°
Use function
sin
cos
cos sin
Error in angle
0f2
0'.'29
0"2
0:'29
0'.'2
Use function
tan
cot
cot tan
Error in angle
0?014
Of 007
Of 01 4
Of 007
Of 01 4
546 EXPLANATORY SUPPLEMENT
correctness of the calculations and of the required answer, and the precision of that
answer, are directly under the control of the computer. Illustrations of the use of
these methods are to be found in the preceding sections, particularly in the calcu-
lation of the local circumstances of eclipses (Section 8.23) and the derivation of the
times of moonrise and moonset (Section 9.3).
The availability of high-speed computers has changed the relative importance
to be attached to the various factors entering into a computation. It is no longer
desirable to restrict the amount of calculation to a minimum: many repetitions of
a simple iteration are often preferable to a more sophisticated direct calculation.
In particular, many of the transformations of coordinates arising in astronomy can
be efficiently handled by the direct use of the accurate formulas, instead of by
approximate series expressions. These often involve multiplying a column matrix
representing the direction cosines by a transformation matrix, as is done, for exam-
ple, for the correction for precession and nutation. Close attention to the precision of
computation is generally unnecessary, since many extra figures can be kept without
any extra work.
Although such machines are used for the computation of the data in The As-
tronomical Almanac, they may not be generally available to all users. The previous
remarks on the techniques of computation, therefore, are included to facilitate cir-
cumstances in which desk calculating machines are used.
Computing techniques adapted to electronic hand-held, desktop, or larger com-
puters are given in most modern texts on computing (see Hildebrand, 1987 and
Press et al, 1986).
1 1 ,2 INTERPOLATION AND SUBTABULATION
1 1 .21 Introduction and Notation
The interpolation methods described in this section, together with the accompany-
ing tables, are usually sufficient to interpolate to full precision the ephemerides in
The Astronomical Almanac. Additional notes, formulas and tables are given in the
booklets Interpolation and Allied Tables and Subtabulation and in many textbooks
on numerical analysis. It is recommended that interpolated values of the Moons
right ascension, declination, and horizontal parallax be derived from the daily poly-
nomial coefficients that are provided for this purpose in The Astronomical Almanac.
The term fp denotes the value of the function fit) at the time t = t0 +ph, where
h is the interval of tabulation, to is a tabular argument, and p = (t-ta)/his known
as the interpolating factor. The notation for the differences of the tabular values is
shown in Table 11.21.1; it is derived from the use of the central-difference operator
i5, which is defined by:
vfp = fp+m-fP-vi- (11.21-D
11 / COMPUTATIONAL TECHNIQUES
547
Table 11.21.1
Differences in Tabular Arguments
Argument Function Difference
1st
2nd
3rd
4th
f-2
f-2
61/2
=/1 -
fo
6-312
= 61/2
- fi_i/2
f-1
f-1
*2-1
*4-i
=/1 -
2/0+/-1
6-i/2
^-1,2
= /2-
/1 -/o+/-1
fo
fo
= *?-
$m
4
= /2-
3/1 + 3/0
Ui
n
%
= sl2
"—US
6312
4
= fz-
4/i +6/0-4/_i
+ /-2
Us
f+2
4
= h-
3/2 + 2/,+2/c-
3/-1 +/_2
The symbol for the function is usually omitted in the notation for the differences.
Tables are given for use with Bessel's interpolation formula for p in the range 0 to
+1. The differences may be expressed in terms of function values for convenience in
the use of programmable calculators or computers.
p = the interpolating factor = (t — to) I (?i — fo) = (f — fo) / h (11 .21-2)
1 1 .22 Interpolation Formulas
Bessel's interpolation formula in this notation is
fP = fo +p6vi + Bi{8l + 6*) + B36l2 + B^4 +1}) + -., (M -22-1)
where
B2=p(p- l)/4,
B3=p(p-l)(p-^) /6,
B4 = (p+l)p(p- I)(p-2)/48. (11.22-2)
The maximum contribution to the truncation error of fp, for 0 < p < 1, from
neglecting each order of difference is less than 0.5 in the unit of the end figure of
the tabular function if S2 < 4, 63 < 60, 64 < 20, and <55 < 500.
548 EXPLANATORY SUPPLEMENT
Everett's interpolation formula in this notation is
fP=fo+ pSm + E20620 +E]S2l+E40640 +E*64l+-, (II .22-3)
where E\ = —p(p — l)(p - 2) / 6,
E2 = (p+l)p(p-l)/6,
4 = -(p+l)p(p~l)(p-2)(p-3)l\20,
E4 = (p + 2)(p+l)p(p- l)(p-2)/120. (11.22-4)
LaGrange's interpolation formula to second order in this notation is
/> = /-! Qp2-^)+/o(1-P2) + /i (\p2 + \p)- (H.22-5)
This formula has the advantage that the interpolation can be done entirely in terms
of the functional values. The disadvantages are that the coefficients are required
to the same number of significant figures as the function, and that the number of
values of the function to use in any particular case is not easily determined without
knowing the size of the differences.
1 1 .23 Inverse Interpolation
Inverse interpolation to derive the interpolating factor p, and hence the time, for
which the function takes a specified value fp is carried out by successive approxi-
mations. The first estimate p\ is obtained from
P\ = (fP ~ fotfm- (11.23-1)
This value of p is used to obtain an estimate of B2, from the critical table or
otherwise, and hence an improved estimate of p from
p = Py~B2{6l + 6])6ll2. (11.23-2)
This last step is repeated until there is no further change in B2 or p; the effects of
higher-order differences may be taken into account in this step.
11.24 Polynomial Representations
It is sometimes convenient to construct a simple polynomial representation in the
form
fp = a0 + a1p + a2p2 + a3ps + a4p4 + ■•• , (II .24-1 )
11 / COMPUTATIONAL TECHNIQUES
which may be evaluated in the nested form
549
fp = (((a4p + ai)p + ai)p + ai)p + ao (1 1.24-2)
Expressions for the coefficients a0> <J|> etc., may be obtained from Stirling's inter-
polation formula, neglecting fifth-order differences:
ao = fo,
a\ = 0m +6-1/2)/ 2 -03,
a2 = &lll-aA, (11.24-3)
a3 = (6l2 + 6ll/2)/l2,
a4 = 6AQl 24.
This is suitable for use in the range -1 / 2 < p < +1 / 2, and it may be adequate
in the range —2<p<2, but it should not normally be used outside this range.
Techniques are available in the literature for obtaining polynomial representation
that give smaller errors over similar or larger intervals. The coefficients may be
expressed in terms of function values rather than differences.
Economized polynomials, such as Chebyshev, provide efficient and accurate
expressions that may be easily evaluated with a small computer. The coefficients a,
of the Chebyshev expansion
n
f*= f + X^'W (11.24-4)
1=1
are computed for prescribed time spans, where fx is the function represented, T/(x) is
the Chebyshev polynomial of the /-th degree, and x is the normalized time variable.
To evaluate a Chebyshev expansion, it is necessary to normalize the time variable
on the interval for which the series is valid. Attempts to use these series beyond
the specified time intervals for which they are valid will produce erroneous results.
An application of Chebyshev polynomials may be found in Section 5.5. Further
information is given in Lanczos (1956) and Fox and Parker (1972).
1 1 .3 PLANE AND SPHERICAL TRIGONOMETRY
In astronomical problems, the sides and angles of triangles on the celestial sphere
may be of any magnitude, and it is undesirable to solve them by methods that
restrict the sides to arcs less than 180°. The general formulas given in Table 11.3.1
are valid for triangles with sides of any length, and may be applied immediately to
Table 11.3.1
Formulas for Plane and Spherical Triangles
(The angles of the triangle are denoted by A, B, C; the opposite sides, by a, b, c. Other formulas may
be obtained by cyclic changes of A, 8, C and a, b, c.)
Plane Triangle Spherical Triangle
a sin B =
bsin A
sin a sin 8 =
sin b sin /4
c - b cos ,4
sin a cos 8 =
a2 =
b2 + C2 - 2 be cos A cosa =
cos b cos c + si n b si n c cos /4
flfrtc C —
a t>w w —
b - ccos A
sin a cos C =
sin b cos c — cos b sin c cos .4
a sin C =
csin A
sin a sin C =
sin csin A
sjn 1(8 + C) =
cos \ A
cos la sin |(B + C) =
cos |/4cos g{b — c)
cos 1(8+ C) =
sin 1/4
cos 5 acos ^(8+ C) =
sin A A cos i(to + c)
asinl(B- C) =
(b — c) cos 1/4
sin ^asin 5(6 - C) =
cos ^ Asin l(b - c)
acos 1(8 - C) =
(b + c)sin 1/4
sinlacosl(8- C) =
sin 1/4 sin |(b+ c)
s =
(a+b+c)
s =
\{a + b + c)
r2 =
(s - a)(-b)(s -
c) / s m2 =
sin(s - a) sin(s - b) sin(s - c) sin s
Area =
sr = |bc sin A
Area =
A + B+ C- 180°
tocsin2 lyA =
(s-b)(s-c)
sin bsin csin2 \ A =
sin(s - b)sin(s - c)
be cos2 j/4 =
s(s - a)
sinbsinccos2 1/4 =
sin ssin(s — a)
tan \A =
r/(s- a)
tan 1/4 =
m / sin(s — a)
Additional Formulas Right-angled Triangle: A = 90°
sin A sin b =
sin Ssina
sin a sin 8 =
sin b
sin acos 8 =
cos b sin c
cos a =
cos b cos c
sin acos C =
sin bcos c
sin A sin c =
sin Csin a
sin a sin C =
sin c
cos a cos 8 =
sin a cot c - sin Scot C
sin B cos a =
cos bcos C
cos acos C =
sin a cot b - sin Ccot 8
cos 8 =
cos b sin C
cos b cos ,4 =
sin b cot c — sin A cot C
sin 8 cos c =
cos C
cose cos A -
sine cot b - sin A cot 8
sin 8 sin c =
sin bsin C
S =
1(A + S+ C)
tana =
tan b sec C
=
- cos(S - A) cos(S - 8) cos(S - C) 1 cos S
tana =
tan c sec 8
sin B sin Csin2 4a =
- cos Scos(S - A)
cos a =
cot B cot C
sin 8 sin ccos2 \ a =
cos(S- B)cos(S- C)
tan b =
sin ctan B
tan |a =
cos(S- A) / M
tan c =
sin btan C
11 / COMPUTATIONAL TECHNIQUES
551
Table 11.3.1, continued
Formulas for Plane and Spherical Triangles
Quadrantal Triangle: a - 90°
sinAsinb = sinS sin bsin A = sin B tan A = -tan Ssecc
sin Acos b = cos Bsin C sin bcos A= -cosBcosc tan A = -tan Csecb
cosA = — cos Bcos C cosb = cosBsinc cosA = -cotbeote
sin A cose = sin Bcos C sin b cos C = cos c tan 8 = sin Ctan b
sin Asin c = sin C sin b sin C = sin Bsin c tan c = sin Btan c
Spherical Triangle in Which b is Small
a - c = -bcos A+ It^cotcsin2 A + -■•
Bsin C = +bA+ \ t£ cote sin 2A + •■■
180° - C — A = +bcot csin A + 1^(1 + 2 cot2 c) sin 2A + ■•■
the general triangles of spherical astronomy without any restriction to values less
than 180°. Furthermore, to make the solution determinate, it is only necessary to
find the algebraic sign of both the sine and the cosine of each arc or angle that may
exceed 180°, to fix the quadrants in which they lie. Any of the cases of the general
triangle is determinate when, in addition to the three given parts, the algebraic sign
of the sine or the cosine of one of the required parts is also given. In most practical
problems it happens that the conditions of the problem supply this sign. In general
triangles the utmost care should be taken to specify unambiguously the direction
of measurement of angles and arcs.
552 EXPLANATORY SUPPLEMENT
In Figure 11.3.1 the arc P'P subtends the small angle b, measured in radians,
r is the radius and P'R is perpendicular to OP at R. The radius of the circle times
the angle subtended in radians equals the arc length, hence rxb = P'P.
When b is small enough that P'R approximates P'P then sinb = P'R / r and
b = P'P I r. Thus, sinfc = b and tanfc = b. If in the small angle of b radians there are
q minutes of arc {q1), then s'mq' = qsin l".
1 1 .4 MATRIX AND VECTOR TECHNIQUES
1 1 .41 Rotation of Axes Using Matrices
In practical computations it is often necessary to change the orientation of reference
frames. Given a right-handed system of rectangular coordinates x, y, z, and rotating
the system positively through the angle i about the x-axis, the new coordinates x' ,
Z1 are related to the original coordinates by
X =y,
y = ycos/ + zsin/,
z = —y sin / + z cos i.
In matrix notation this becomes (x',y',z') = (x,y,z)P(i), where
(11.41-1)
P(i) =
10 0
0 cos / — sin /
0 sin / cos /
(I 1.41-2)
Similarly, for a positive rotation j about the y-axis, (x',y',z') = (x,y,z)Q(j), where
Q(j) =
cos j 0 sin j
0 1 0
- sin y 0 cos j
(11.41-3)
Again, for a positive rotation k about the z-axis, (x'./.z') = (x,y,z)R(k), where
R(k) =
cos k — sin k 0
sin* cosk 0
0 0 1
(11.41-4)
The equation for two positive rotations may be written as
{x',y!,z') = (x,y,z)P{i)Qij),
(11.41-5)
11 / COMPUTATIONAL TECHNIQUES
where the matrix product is
553
P(')Q(J) =
cosy
sin i sin j
— cos /' sin
0 sin j
cos/ — sin /'cosy
sin / cos i cos /
(11.41-6)
For three successive positive rotations (j), (/), (k) about the jc-, y-, and z-axes
respectively, (x',y',z') = (x,y,z)P(i)Q(J)R(k), and the matrix product is
cosy cosk - cos y sin k siny
P(0Q(j)R(k) - sin i siny cos k + cos i sin k - sin i sin j sin k + cos i cos k - sin i cosy
_-- cos i siny cos k + sin /sin* cos /siny sin k + sin /cos k cos /cosy
(11.41-7)
This case comprises a rotation (/) about the x-axis, a rotation (j) around the
new position of the y-axis, and a rotation (k) around the final position of the z-axis.
The matrix denoting any number of rotations can be derived by multiplication
following the precepts below.
Two matrices A and B form the product AB, where the superscript indicates
the row and the subscript indicates the column,
A =
r
a\
-b\
b\
*il
4
B =
b\
H
4
°i -
A
b\
H.
(U.41-8)
AB =
■a\b\+a\b2x+a\b\ a\b\+ a\b\ + a\b\ a\b\ + a\b\ + a\b\
a]b\ + a\b] + a\b] a\b[ + a\b\ + a]b\ a\b\ + a\b\ + a\b]
a\b\ + a\b\ + a\b\ a]b\ + a\b\ + a\b\ M + <M + a\b\
•2"\
J3WI
'I "2
2u2 ■
*3"2
'1"3
(11.41-9)
1 1 .42 Spherical Coordinates Using Vectors
Consider a triangle ABC (Figure 11.1.1) scribed on the surface of a sphere of unit
radius whose center is denoted by O. Each side abc is a portion of a great circle
defined by BOC = a, AOC = b, and AOB = c.
A is the angle between the planes BOA and COA, B is the angle between the
planes AOB and COB, and C is the angle between planes AOC and BOC.
Take I, m, n. as nonorthogonal unit vectors along OA, OB, and OC respectively.
The cross product 1 x m is a vector of magnitude sin c directed ort hogonally to the
plane AOB. In the same manner, the cross product 1 x n is a vector of magnitude
sinb orthogonal to AOC. A is the angle between these two vectors.
554
The dot product of these two vectors is
EXPLANATORY SUPPLEMENT
(1 x m) ■ (1 x n) = sincsinfccosA
= 1 • [m x (I x n>]
= 1 [I(m ii) n(l m)|
= (m - n) - (1 • n)(l ■ m)
= cos a - cos b cos c, (11.42-1)
which gives
cosa - cosi?cosc + sinbsinccosA. (11.42-2)
Similar equations for cosb and cose are easily obtained.
From the definition of the vector product, we find
, |(1 x m) x (1 x n)
sinA =
I x m| |1 x n|
-l[m, l,n] + m[l,l,n]
sin b sine
0, m, n]
sin b sin c
Thus,
sin A sin/? sinC
sin a sin£> sine
(11.42-3)
(11.42^4)
Other formulas of spherical trigonometry (see Table 11.3.1) may be obtained
from the two above.
Another application of vectors to the solution of a problem in spherical trigonom-
etry concerns transforming the coordinates of a star from right ascension (a) and
declination (6) to ecliptic longitude (A) and latitude (j3).
Given the triangle EQS (Figure 11.42.1) on a sphere of unit radius with center
at O, Q is the pole of the celestial equator, E is the pole of the ecliptic, and S is the
star.
The side EQ is the obliquity of the ecliptic (e), the side QS is 90° - <5, and the
side ES is 90° -0. Two great circles passing through E and Q, each being orthogonal
to side ES, intersect the equator and ecliptic planes at the vernal equinox. Then the
angle EQS = 90° + q, and the angle SEQ = 90° - A.
Using these quantities the problem is easily solved by the equations previously
derived:
sin 0 = sin 6 cos e — cos <5 sin a sin e,
cos 0 cos A = cos b cos Q,
cos 0 sin A = sin 6 sin e + cos 6 sin a cos e. (11 .42-5)
11 / COMPUTATIONAL TECHNIQUES
555
Q
s
Figure 11.42.1
Triangle on unit sphere
The problem can also be solved by the use of vectors. Assume axes I. m, n
(forming the conventional right-handed triad) with 1 pointing to the vernal equinox
and n directed toward the north pole of the equator (Q). Also assume the right-
handed triad u, v, w with u directed to the vernal equinox and w toward the pole
of the ecliptic (£).
Then
u = l
v = m cos e - n sin e
w = m sin e + n cos f . (11 .42-6)
The components of OS projected onto the equatorial and ecliptic axes are
cos 6 cos ct, cos b sin ct, sin 6,
cos 0 cos A, cos 0 sin A sin/3,
leading to the original trigonometric equations.
1 1 .43 Specific Coordinate Transformations
For geocentric spherical coordinates there are thus the four practical reference sys-
tems of:
556 EXPLANATORY SUPPLEMENT
(1) Azimuth (A) is measured from the north through east in the plane of the
horizon, and altitude (a) measured perpendicular to the horizon. In as-
tronomy, the zenith distance (z - 90° — a) is more generally used, but the
altitude is retained in the formulas for reasons of symmetry.
(2) Hour angle (h) is measured westward in the plane of the equator from the
meridian, and declination (6) is measured perpendicular to the equator,
positive to the north.
(3) Right ascension (a) is measured from the equinox eastward in the plane of
the equator, and declination (<5) is measured perpendicular to the equator,
positive to the north.
(4) Longitude (A) is measured from the equinox eastward in the plane of the
ecliptic, and latitude {(3) is measured perpendicular to the ecliptic, positive
to the north.
Formulas for transforming equatorial right ascension and declination to ecliptic
longitude and latitude may be written using direction cosines (unit vectors) and
(rotation matrices) as follows:
" cos (i cos A "
"1
0 0 "
cos 8 cos a '
cos 0 sin A
0
cos t sin e
cos 8 sin a
(11.43-1)
sin/3
.0
— sin e cos e
sin 6
The inverse transformation, using [Re] to denote the rotation matrix, is
'cos (3 cos A"
"cos<5 cos q"
cos (3 sin A
= [R,rl
cos 6 sin a
sin j3
sin 8
The transformations correspond to simple rotations about the jc-axis through
e, which is the obliquity of the ecliptic corresponding to the particular equator and
ecliptic used.
Transformation from right ascension/declination to local hour angle/declination
may be carried out by substitution of
h — local sidereal time — a.
The two systems are identical apart from the origin and direction angular mea-
surement. The transformation may also be considered as a rotation about the axis
perpendicular to the equatorial plane and positive in the direction of the North
Pole. Thus,
cos 6 cos h '
costs sin ;,s
0"
cos 8 cos a.
cos 8 sin h
— sin fj cos tg
0
cos 8 sin a
(11.43-3)
sin 6
0 0
1
sin 5
11 / COMPUTATIONAL TECHNIQUES 557
where tf is the local sidereal time.
Transformation of local hour angle/declination to azimuth/altitude involves a
rotation of the reference frame through an angle 90° — <j> in the plane of the. meridian,
where 4> is the geocentric latitude, followed by a rotation about the new vertical by
180°. Thus,
"cos a cos A "
cos a sin A
sin a
-1 0 0
0-10
0 0 1
sin (j) 0 - cos 4
0 1 0
cos tfi 0 sin 4>
cos 6 cos h "
cos 6 sin h
sin 6
(11.43-4)
In this transformation the diagonal matrix is its own inverse and, if [R^] denotes
the rotation in the meridian plane with inverse, then
simp 0 cos i
0 1 0
— cos d> 0 sin t
(11.43-5)
and
"cos 8 cos h
cos 6 sin h
sin (•■
-1 0
0 -1
0 0
cos a cos A '
cos a sin A
sin a
(11.43-6)
The corresponding equatorial rectangular coordinates and distance are denoted
by X, Y, Z, and R for the Sun and by £, rj, (, and A for the planets; they are derived
from the spherical coordinates by the formulas:
x t x
— or — = cos o cos a,
R A
Y 7? * -
— or — = cos o sin a,
R A
— or ^=sin6. (11.43-7)
R A
Geocentric ecliptic rectangular coordinates are rarely (if ever) used.
For heliocentric coordinates there are only the two practical reference systems
the equatorial and the ecliptic. In the equatorial system only rectangular coor-
dinates are used. The relationships between the ecliptic rectangular coordinates
(Xc,yc,Zc), the ecliptic longitude, latitude, and distance (l,b,r), and the equatorial
rectangular coordinates (x,y, z) are
- r cos b cos I
X
yc -
- r cos b sin/
= m
y
.Zc -
= r sin b
_z_
558
EXPLANATORY SUPPLEMENT
.V
'xc~
(11.43-9)
The conversion from heliocentric to geocentric coordinates is performed in terms
of equatorial rectangular coordinates using
ri = v + Y,
(11.43-10)
where X, Y, Z are the geocentric coordinates of the Sun.
The calculation of the spherical coordinates from the rectangular coordinates,
or from the known direction cosines, typified by
A cos 6 cos a = £,
A cos 6 sin a = i],
A sin 6 = C>
(11.43-11)
is performed by
tan a = rj I £,
zi = (f2 + /r + C2),
cot a = £ / jj,
sin<5 = C/A (11.43-12)
The quadrant of a is determined by the
of £. Zi and Zlcos<5 are always positive,
as
q = tan-1 / £,
= cot"1
5 = sin"'C/A
igns of £ and r?, and that, of 6 by the sign
rhe formulas for a and 6 may be written
or arctan;j/£,
or arccot £ / //.
or arcsinC/A (11.43-13)
provided that the appropriate values, and not necessarily the principal values, of
the multivalued functions are taken.
Many of the conversions above correspond to a simple rotation of the frame of
reference about one of its axes. These are special cases of the general conversion
11 I COMPUTATIONAL TECHNIQUES 559
from a set of axes designated by x, % z to a set designated by x1, /, z'. The two
systems are connected by the formulas
x = l]x' + i2y' + hz', x' = lix + miy + niz,
y = m\xl + m2y' + m^z' , y' = l2x + m2y + n2z,
z = n\x' +n2y' + mz', z' = I3X + m3y + n3z, (11.43-14)
where l\,m\,n\\ l2, m2, n2; and I3, m3, /13 are the direction cosines of x' , y', z' referred
to the system x, y, z. The direction cosines satisfy the relations typified by
l\ + m\ + n\ = 1 ,
l2h + m2mi + n2ni = 0,
1} + + = I,
tftiB] + m2n2 +m3«3 = 0. (11.43-15)
These nine quantities can be expressed in terms of the Eulerian angles 9, <p, tp
as
1 1 = + cos 4> cos 9 cos i> - sin <j> sin ip,
h = — cos <p cos f? sin i/j — sin 0 cos ip,
= + cos <p sin 9,
m\ = + sin <p cos 9 cos tp + cos 4> sin tp,
m2 = - sin <p cos 9 sin rp + cos <p cos rp,
W3 = + sin <p sin 9,
«i = — sin # cos
n2 = + sin 9 sin xp,
n3=+cos0. (11.43-16)
In this case the conversion corresponds to a rotation cp about the z-axis, 9 about
the new position of the y-axis, and ip about the new (and final) position of the z-
axis. The transformation is equivalent to a single rotation about some line not in
general coincident with one of the axes; but such single rotations are not frequently
encountered in astronomical practice.
560
11.5 NUMERICAL CALCULUS
EXPLANATORY SUPPLEMENT
11.51 Numerical Differentiation
Successive derivatives of a function are closely related to the successive orders of
differences. Thus to obtain the exact expression for the derivatives in terms of
differences it is necessary merely to differentiate any of the interpolation formulas
(see Section 11.2) as many times as required.
Differentiation of Bessel's interpolation formula leads to
hfp = 6y2+l- (p-^j(620 + 6^+^(1 -6p + 6P2)8l1 + :., (11.51-1)
where f'p denotes the value of df I dt at the point t = to + ph.
This formula is intended for use in the range 0 < p < 1; the maximum value of
the third-difference coefficient is 1/12 and occurs at p = 0 and 1. At p = 1/2 the
formula reduces to
hfm = 6l/2 - ^S\a + ^65m - - . (1 1.51-2)
Differentiation of Stirling's interpolation formula leads to
hfp = v60+p6t- g(l -3p2)»6l + -, (11.51-3)
where, for example, p,6^ = j(^^,/2 + 5f/2). This formula is intended for use in the
range —l/2<p< 1/2; the maximum value of the coefficient of the mean third
difference is 1 / 6 and occurs at p - 0. At p = 0 the formula reduces to
hf0 = p6o - pi&l + ~n65Q - (11.51-4)
The condition for a maximum or minimum is that f'p = 0. For Stirling's formula
this condition may be expressed as:
pa-^+i(i-y^-. (]]51_5)
This equation must normally be solved by successive approximations. The maxi-
mum contribution from the third difference is 1 / 6 p.6\ I 6%, and if this is negligible
p may be evaluated directly from
p.-Ufc/jg*-! -6-1/2/4 (11.51-6)
11 / COMPUTATIONAL TECHNIQUES
561
Table 11.51.1
Derivatives to an Order of 10
Derivatives at a Tabular Point Derivatives at a Halfway Point
= f *0
~g"*0
+ - -
M'V2
= *1/2
1 *3
_ 53*1/2
+
3 *5
535 *1/2
='o2
1 X*
- T5*0
.4. 1 A6
+ 50*0
■ M*?/2
- S?M*1/2
+
5760^*1/2 -
-1^1
+ TSJ'^O
" *1/2
_ 5*1/2
+
37 ,-7
T920*1/2 -
6*0
+ 535*0
= ^s1/2
53^*1/2
= M«f
- w
" *1/2
53*1/2
+
47 <9
1152*1/2 ~
- *0
-H
. 13 i10
+ 535*0 '
" J 1/2
= V-l>%2
+
209 ,.c10
1555^*1/2 -
*7/S"
- M
h7fvil
= *~1/2
+
133 c11
1920*1/2 ~~
»8/q™
3*0
4. 31 A12
+ 555*0
= rf'a
11 „/,10
+
5760*'*1/2
= W
1 i.Jt11
+ •■■
-4
3 ,11
_ 5*1/2
+
=c
12*0
+ ••■
"1X
13 „i12
+
In astronomical usage, the terms variation and motion are synonymous with
"derivative with respect to time," when qualified by an adjective defining the unit
of time, and are usually evaluated for the tabular points. The term secular variation
usually implies a second derivative with respect to time.
The second derivative fp is obtained by differentiating the formulas for f for
Stirling's formula this leads to
h1f'p = 6l+pn6l + -, (11.51-7)
and, when p = 0, to
h2f; = S20-^0 + ^660--. (11.51-8)
In terms of central differences, the derivatives to order 10 at the tabular point
and at the half-way point are listed in Table 11.51.1. Primes are used to denote
differentiation with respect to x. For example,
h2fp = h2 £j/(.*„ + ph) = ^fp. ( 1 1 .5 1-9)
For higher derivatives the primes are replaced by italicized roman numerals. In
general formulas, fptt) is used to denote the «th derivative of / with respect to x at
the point Xo+ph.
562 EXPLANATORY SUPPLEMENT
The formulas in Table 11.51.1 may also be written by substituting for the dif-
ferences their values in terms of the function. Additional expressions for derivatives
in terms of the functional values are as follows:
hfQ = \(j\ -/-,)- ^3/o + -
- — C-/2 + 8/, - 8/_, + f-2) + ^65fo ~ -
= ^(/.i - 9/2+45/, - 45/_j +9/_2 - f-3) - 7^7o + -
*2/o = C/i - 2/o +/-])- ^<54/o + -
= ~{-h + 16/i - 30/o + 16/--, - f-2) + ^6/o - -
= ^(2/3 - 27/3 + 270/, - 490/c + 270/_, - 27/_2 + 2/_3) - ^58/o + -
&/w = Cfi -h)-~63fl/2 + -
= + 27/, - 27/0 + /-,)+ jj^/ia - -
24 640
= i^q(9/3 - 125/2 + 2250/, - 2250/o + 125/_, - 9/_2) - ^S7fV2 + -
t>2fm = ~ fi ~ fa +/-.) - J^Vi/2 + -
1 259 *
= 4^(-5/3 + 39/2 - 34/, - 34/0 + 39/_, - 5/_2) + ^M6/i/i
(11.51-10)
For further formulas see the Interpolation and Allied Tables and the Handbook
of Mathematical Functions.
1 1 .52 Numerical Integration
When studying motions within the solar system it is necessary to use the solution
of ordinary differential equations of the second order. This normally involves the
simultaneous integration of three second-order differential equations to produce an
ephemeris in rectangular coordinates of a planet, comet, or satellite with time as
the independent argument.
In numerical differentiation, the derivatives of a function are expressed in terms
of the differences. In numerical integration, the opposite method is used. Derivatives
of some function are computed and checked by differencing, to form a table of /,
8, 6 , £3, S4, and so on. In the simplest case, where the values of the function to be
integrated can be calculated in advance, it is possible to compute a starting value
11 / COMPUTATIONAL TECHNIQUES
563
of the first summation /' and to extend the table in either direction by applying
successive values of /. For a double integration, a starting value for /" must be
evaluated and successive values of /' applied. Expressions to evaluate the first and
second integrals are derived by integrating an interpolation formula, for example,
f(x) dx = co
f(x)dx1 = u?
Ml
/o + i2^°
17
5760
1
'1/2
367
967680
Jl/2
240
31
60480
(11.52-1)
Additional formulas are given in Interpolation and Allied Tables and in Handbook
of Mathematical Functions.
In the integration of orbits the problem is more complicated because the func-
tion that is to be integrated is undetermined until the integral becomes known. Thus
it is necessary to proceed stepwise using extrapolation and approximation at each
step. Because of the availability of computing machines, as well as the rectangular
coordinates of the principal planets, two methods for the numerical integration of
the orbits have proved useful, Cowell's and Encke's.
When one is utilizing Cowell's approach, the equations of motion are integrated
to give directly the positions of the perturbed body in rectangular coordinates. For
a planet or comet, the origin usually is the Sun and at each step the attractions of
the perturbing bodies would be added to that of the Sun.
With Cowell's method, however, the origin may be any perturbing body, or even
a double star, since no use is made of a conic section as the first approximation. The
single condition is that the motions of all perturbing bodies relative to the selected
origin should be known. A disadvantage of Cowell's method is that the integrals
comprise numerous significant figures that vary extensively with time. Thus, the
integration tables are prone to converge rather slowly, which may require the use
of a short step interval. The basic equations in the Cowell approach are
with similar equations in y and z. The masses are expressed in terms of the mass ot
the Sun, k equals the Gaussian constant, m is the mass of the perturbed body, and
mi the masses of the perturbing bodies. The first term is the attraction of the Sun
on m, and the first term in parentheses is the attraction of irtj on m. The last term
in parentheses is the action of the perturbing bodies on the Sun.
564 EXPLANATORY SUPPLEMENT
In Encke's method the unperturbed orbit of the mass m about the Sun is
assumed known. Hence only the difference between the unperturbed position and
the perturbed position needs to be determined. The equations may be written
*~*>-t'*0 (I - -|)*E*S ( ^ - |) ' (H.52-3)
with similar equations for 77 and £ These equations could be integrated directly,
but since f is small, many figures would have to be carried in the term (xq/ r^-xl V3)
to retain significance in the difference. On the assumption that f, n and C are so
small compared to x, y, z that their squares can be neglected, the equation can be
transformed by the use of auxiliary quantities q and / where
-*o£ + ym + zoC
* 7 9
f= • (11.52-4)
The final Encke equations become
W2£ = or^fl + m)±(fqx + E ^k2mj - I . (J 1.52-5)
Where w represents the interval of integration. Similar equations hold for 77 and f.
Encke's procedure normally allows a larger step size than Cowell's but when
close approaches happen, the Encke perturbations tend to grow rapidly in size,
requiring a smaller step interval.
Other approaches to the numerical solution of differential equations are pro-
vided by Runge-Kutta methods. These are single-step methods with each step being
independent of those preceding.
Given:
dx ,
— = f(t,x); x(t0) s= x0. (1 1.52-6)
A widely used Runge-Kutta method (to the fourth order) involves the following
operations for one step. Here h is the step interval.
fo = f(k>x),
h = /(^0+|M) + |A/o) ,
f2=f(t0+^h,x0 + ^hflSj ,
11 / COMPUTATIONAL TECHNIQUES
h =f(to + h,x0 + hf2),
565
(11.52-7)
Then
x(t0 + h) = x, = x0 + l-(f0 + 2/, + h + f3),
6
Similar equations exist in y and z.
Equations to the fourth order are not sufficiently accurately for most orbital
calculations but this serves as an example of the method. For modifications of the
method refer to Danby [Section 11.7].
In all the step-by-step methods for numerical solution of differential equations
some precautions apply:
(1) The interval of tabulation must be monitored in cases of close approach,
since the approximations used in the iterative process may prove to be of
insufficient accuracy. Automatic methods to vary the step size are now in
use with some methods.
(2) The accumulation of round-off errors must be monitored. The probable
error after n steps can be estimated.
(3) If the method of solution adopted makes use of truncated formulas either
corrections must be applied subsequently or the number of figures in the
solution must be reduced accordingly.
Examples of numerical integrations may be found in Brouwer and Clemence
(1961) in Astronomical Papers American Ephemerides, vols. XII and XX, pt. 1,
and in Danby (1988) (Section 11.7).
An example of a multistep scheme is the Adams-Bashforth method. The order
of this method is limited by the number of terms the user is willing to compute.
The formulas for this method are
v
xn+] =Xn+hY^ akVkfn, (1 1.52-8)
where
the number of terms,
1 5 3 251 95
'2' 12' 8' 720' 288'""
/„,
N =
a =
Vkf„ =
/„ =
566 EXPLANATORY SUPPLEMENT
This method can be started using a single step method such as the Runge-Kutta.
To differentially correct the initial conditions of the motion, one integrates seven
orbits simultaneously. The first orbit uses unaltered initial conditions. Each of the
rest of the orbits uses initial conditions for which one coordinate or velocity has been
incremented by a small amount. Earth based observations are right ascensions and
declinations of the satellite. The residuals for these observations can be expressed
as
eta, . eta, . eta, . eta, , ■ eta, . ■ eta, . ,
dtj 86, ddi 86, . d6j . &6i
A6i = — - AXn + — ^AKo + -^-AZb + -^-AX„ + —t^AYq + -^-AZq,
dX0 OY0 dZo ^ dX0 8Y0 8Zq
(11.52-9)
where a and 5 are the right ascension and declination respectively; Xq, Yq, Zq, Xq,
Yq, and Zq are the initial coordinates and velocities; the subscript i refers to the
number of the observation. The partial derivatives are evaluated numerically, and
can be written in the form
eta, _ ajiXp + AXp, Yu.Zp.Xp, Kp.Zu) - q,(Xq, Yq,Zo,Xq, Yq,Zq)
aXo AX,,
The differential corrections can be written as
AR = (ATArlATAr. (11.52-11)
where AR is the matrix of corrections to the initial conditions, A is the matrix of
the partial derivatives, and Ar is the matrix of residuals.
1 1 .6 STATISTICS
1 1 .61 The Accumulation of Error
The formulas listed in Table 11.61.1 indicate the way in which errors accumulate
in simple arithmetical operations. The distribution of accumulated error is usually
nearly normal, whatever the distributions of the original errors, so the standard
deviation of the error, calculated from the formulas, is of value in the examination
of the differences of tabulated functions. The notation used and assumptions made
in the table are:
a,- is the absolute error of Aj, in the sense of given value minus true value. These
errors are assumed to be unbiased and independent .
11 / COMPUTATIONAL TECHNIQUES
567
Table 11.61.1
Accumulation of Error in Arithmetical Operations
Operation
Absolute Error
Variance of Error (unit 1/12)
Em*
Em «
Em a'
"^Ett miA>
ma JPJ^ m,Ai
E« 'Aj lar9e>
^nPa if q, = q for all i
ka + e Y^l-i ^'a' + e
*f2° + 1E"=,'c?Q' + 1
krA
kra+ e
9(P)A
g(p)a + e
/01{9(P)}2*.« + 1
Em P'A (for <*,= <*)
AfAz
A2a^ + A-\a2 + e
A^ci\ + A^a2 + 1
A"
nA"~ 1 a + e
W-Zq + 1
1
A
(-^)a+e
(1 lA*)a + 1
AV2
nm difference of A(x)
f(A) by direct interpolation
(^)a, -(|}a2 + e
af'(A) + er + e, + e
f[A) by inverse interpolation {a-e,- ei)f'(A) + e
it a is constant
{f'(A)}2a + vr + V, + 1
(f{A)}Ha+Vr + vt) + 1
is the variance, or mean square value, of the error of /4,-. The standard deviation
of the error is the square root of variance.
, m, and n are integers, k is a known constant, r is equally likely to take any value
between -1/2 and +1 12. p is equally likely to take any value between 0 and 1.
568 EXPLANATORY SUPPLEMENT
k, r, and p are considered to be free from error, that is, they are exact or are known
to more significant figures than the quantities A,.
e is the error arising from the rounding of the final result; it is assumed that a
guarding figure is kept in intermediate results, e is equally likely to take any
value between ± 1 / 2 so that its variance is 1/12, equal to the integral of e2
between + 1/2, in units of the last place retained. For convenience all variances
are expressed in this unit of 1 / 12.
In addition, er is the error due to the rounding-off errors in the tabular values,
including their effects on the differences; et is the truncation error, and vr, v, are
the variances of er and e,. If all values of the interpolating factor p are equally
likely, vr varies from 0.67 for linear interpolation to 0.81 for interpolation with fifth
differences. If f(A) is determined from a critical table its error is that of a simple
rounding-off error unless the number of figures in the argument is small or unless
the interval in the respondent is greater than unity.
1 1 .62 The Method of Least Squares
11.621 Introduction When one is measuring the physical universe, certain con-
stants must be determined from observation, such as the orbital elements of a
satellite about its primary or the rotational elements about its axis. Methods for oh
initio determination of orbital elements from three or more observations are well-
documented in the literature (Herget, 1948). The more typical case assumes that
approximate values of the constants are available on which a provisional theory
can be based. This theory is then used to construct positions against which the
observations can be compared. Thus, each observation minus the computed place
forms a difference, referred to as an O — C. The errors in these differences can arise
from three sources: errors in the observations; errors in the constants on which the
computed positions are based; and errors in the theory itself. Only the first two er-
ror sources will be treated in the following discussion, which concerns the analysis
of O — C's to improve the initial approximation to the constants.
11.622 Errors of Observation: Frequency Distribution Errors of observation may
arise from many sources, The size of these errors vary and, in general, follow the
normal distribution curve or probability distribution curve wherein small errors
occur more frequently than large errors, and positive errors occur with the same
frequency as negative errors. In most problems, the frequency distribution of the
errors is not known in advance. However, if each observation is assumed to be af-
fected by the sum of several simultaneous, independent errors, we may assume the
distribution of the errors to be accidental rather than systematic, the latter be-
ing the case where all the errors have a common origin. As there is no satisfactory
11 / COMPUTATIONAL TECHNIQUES 569
mathematical theory of systematic errors, we must proceed initially on the assump-
tion that the errors are accidental. In this case if a set of random variables x are
independent of each other, the theory of probability postulates that the frequency
distribution of the sum of such variables, each with its own frequency distribution,
will slowly approach the function
Y= 4=exp(-/i2*2), (11.622-1)
where h is defined as the index of precision. This is related to the standard deviation
a, or the root-mean- square error by h2 — 1 / 2a2.
The best, fit to a series of n observations, according to the principle of least
squares, is the one that reduces the sum of the squares of the errors to a minimum.
Assuming that the index of precision is constant in this set of observations and
letting the errors associated with each observation, 0,-, be x,, the probability, P,, of
occurrence of error jc,- becomes
h
Pi = —= exp(-h2xf) dxu (11 .622-2)
Assuming that P is independent, the probability that each error will occur in turn
is
P=^JLSj exp^-^J^xfjdxidxn ... dxn. (11.622-3)
Thus it can be stated that the set of errors that maximize P in Equation 11.622-3
will produce the best fit to the observations because at that point ^ii-^f *s a
minimum.
If the index of precision varies with each observation, the introduction of the
factor hi requires knowing the correspondence between the hi and the arbitrary
weight of an observation, based on unit weight. If the precision index corresponding
to weight 1 is h, and the precision index corresponding to weight W\ is h\, the
probability that an error of size x will occur in an observation of weight 1 is
P = 4= exp( -h V) dx. (11 .622^1)
Also the probability of an error of size x occurring in an observation of weight W\ is
P = -75 exp(— h2x2)dx.
(11.622-5)
57Q EXPLANATORY SUPPLEMENT
Assuming that P = P\, that is that an error of size x has an equal probability of
occurring in both Equations 11.622 4 and 11.622-5, leads to
W| = (h\ I h)~,
Wi = (hi/h)2, (11.622-6)
which shows that the weights are proportional to the precision indices. In terms of
the precision index h and the weights Wi,
p = ( — = J h" exp -h 2^
i=i
dxidx2...dxn. (11.622-7)
showing that for P to be a maximum, the Yl"=\ wi *l must be a minimum.
Assume 0, (where f = 1, 2, 3, n) to be observations of a quantity for which r
is the true value. The error of each observation is x, = 0, - r. Using the standard
deviation rr instead of the precision index h, the best estimate 7 of the true value
t is that which reduces
M2a2YSO\ -if
to a minimum. To do this requires that d = 0; i.e., 2£(0; - 7) = 0, from
which 7 ■■=■ 1 / »X)Oj, or the best estimate is the arithmetic mean.
Given n observations 0,, having normally distributed errors xh the error of the
arithmetic mean is
x=-Yxh (11.622-8)
which is a linear combination of n random variables. If the mean error of one
observation is a, then it can be shown that the mean error of the arithmetic mean
is the a of one observation divided by the ^/n.
Once the estimate 7 of the true value r of the measured quantity is obtained,
a residual r can be formed by 0f — 7 a r,. Also Of ~T= #«, where the are the true
errors of the observations. Then
y^Jtj = 2J Oi —jtT a H7 - »T,
7 = T + 2^ x,- / /l,
r, (11.622-9)
Let a, be the mean error of the residuals and ax be the mean error of the errors Xt,
then
a2 = <(«-!)/ n)2a2 +(»-!/ n2)a2 =(«-!/ «)<r2 ( 1 1 .622-10)
11 / COMPUTATIONAL TECHNIQUES 571
Since a2 = ^rin, then cr2 = £V/n - 1 and the mean-square error of the arithmetic
mean in terms of the residuals is
r= y^r2/H(n- 1). (11.622-11)
1 1.623 The Method of Least Squares In many problems the quantity measured is
not the one that we wish to correct. A simple case involves one unknown, to be
determined from measurements Q, where i = 1, 2, 3, n. Thus C = /(")• Let t]c be
the provisionally known value of n, and % be the the value to be determined from
the measurements, then the difference % - nf = A??. Calculating from jfc for each
measurement 1, gives A,< = Q - Q. Using a Taylor's series expansion, AC may be
written as a function of An,
AC = gA„ + ^(A^)2 ^fl(A^)3 + - ■ (11.623-1)
Assuming Ar/ is small enough that its square and higher powers can be ignored,
then
fdf
, A7? = AC (11.623-2)
\dvJ 1
and we have n equations where in each case the (9/ / 9n) is a known numerical
coefficient. Denoting the coefficients by A, the equations of condition can be written
AA77 = AC- The preferred value of the unknown is that for which
9 =5>,Ar,-A,C)2 =0. (11.623-3)
dArj
Differentiating gives £A2An = £AAC, thus An = £AAc/ £A2 and An+n, =
In a more general problem it may be necessary to determine more unknowns
than one from a series of measurements. The constants of a planetary orbit C,
(where i = 1, 2, 3, n) may already be approximately known and can be used to
compute the r)c corresponding to each observation 170, then % - »fc = An. Let the
corrections to the approximate values of the orbital constants be Eh so that the
corrected constants are C, +£, . The equations of condition can then be written
Setting dr]/ dC\ =A, dr\l OC, =B,dv/ dCy = C, and so on, the conditional equation
can be written
A£, + BE2 + CE? + -+ NE„ = An. ( 1 1 -623-5)
572 EXPLANATORY SUPPLEMENT
Supposing there are no errors in the An, i.e., that the theory used is adequate, and
that the observations contain no errors, then the only source of error in the solution
comes from the approximate values of the constants. The theory of linear equations
states that n independent equations of condition are necessary and sufficient to
determine the Et. With the above restrictions as to the source of errors, the solution
will yield values of E that will exactly satisfy the n equations. Normally, however,
accidental errors in the values of An prevent the derivation of the exact values of
Ej. Increasing the number m of observations and thus the number of conditional
equations will decrease the effect of the accidental errors. In some problems in
dynamical astronomy it is not unusual to have m exceed n by a factor of 1000.
Since values of £/ are derived that do not exactly satisfy the conditional equations,
the procedure is to evaluate the left-hand side of the equations and subtract from
the corresponding An to produce residuals r. The best values of the £, are those
that reduce the sum of the squares of the residuals (r) to a minimum as discussed
earlier.
In the instance where there are m equations of condition, with m > n, the
procedure is to form n normal equations for solution. The normals are formed as
follows. Given the conditional equations,
A\E\ + BlE2 + ClE3 + -~+N]E„ = Ani,
A2E] + B2E2 + C2£3 + •■■ + N2E„ = A772,
A3Ei + B3E2 + C3E3 + ■•■ +N3En = An3, etc., (11.623-6)
multiply through each conditional equation by the coefficient A,- and sum to form
the first normal equation in the array that follows, where the brackets represent
summation. Then multiply through each conditional equation by B< and sum to
form the second line of the array. Repeat with C, to form the third line of the array.
The procedure is repeated through the coefficient of En and the symmetry of the
resulting array of n equations in n unknowns reduces the complexity of the solution.
[Af]Ei + [AiBi]E2 + [AiQ}E3 + ■■■ + [AtNi]En
[BiAj]E] + [Bj]E2 + [5;C,]£3 + - + [B,A/,]£„
[dA^ + [mm + [cj]E3 + - + [cmm*
= [AArji],
= [BiAml
= tCiAml (11-623-7)
11 / COMPUTATIONAL TECHNIQUES 573
When the m equations of condition are not all of equal weight, each equation may
be multiplied through by the square root of its weight before forming the normals.
Methods of solution of the normal equations are numerous and are detailed in
the references given at the end of this chapter.
An example of the method of least squares is the problem of fitting a straight
line through a series of measured points, where the function A + Br) - f(rj) is real
and continuous.
Example 1 The data are
7? I 0 0.80 1.60 2.40 3.20 4.00
f(rj) I 1.01 3.87 6.52 9.36 12.07 14.87
The coefficients and right hand members of the six equations of condition can be written
in tabular form:
A
B
f(v)
1
0
1.01
1
0.80
3.87
1
1.60
6.52
1
2.40
9.36
4
3.20
12.07
1
4.00
14.87
Assuming all of the equations are of equal weight, the normal equations are
6A+ 125 = 47.70
124 + 35.25= 134.096, (11.623-8)
and the solution given by the least squares approximation is
A = +1.04
B = +3.455
or
f(rj)"z(v)= 1.04+3.455r?. (11.623-9)
The values derived from the solution compared to the measurements are
r) 0 0.80 1.60 2.40 3.20 4.00
/(??) 1.01 3.87 6.52 9.36 12.07 14.87
2(77) 1.04 3.77 6.54 9.30 12.07 14.83
r .03 .10 .02 .06 .00 .04
The sum of the squares of the residuals is 0.0165 and the root mean square error of the
six data points is 0.052.
574
EXPLANATORY SUPPLEMENT
11.7 REFERENCES
Abrarnowitz, M, and Stegun, I. (1973). Handbook of Mathematical Functions, 9th ed.
(Dover Press, New York).
Bowditch, N. (1981). Section on mathematics. American Practical Navigator (U.S. Gov-
ernment Printing Office, Washington, DC) pp. 436-441.
Brouwer, D. and Clemence, G.M. (1961). Methods of Celestial Mechanics (Academic Press,
New York).
Chauvenet, W. (1850). Treatise on Plane and Spherical Trigonometry (H. Perkins, Philadel-
phia) .
Chauvenet, W. (1892). A Manual of Spherical and Practical Astronomy Philadelphia, 5th
ed. (Constable and Company Limited, London) (Reprinted Dover Press, New York,
1960).
Danby, J.M.A. (1988). Fundamentals of Celestial Mechanics 2d ed. (Willmann-Bell, Rich-
mond, VA).
Duncombe, R.L. (1969). "Coordinates of Ceres, Pallas, Juno and Vesta 1928-2000" Astro.
Papers Amer. Eph, XX, II (U.S. Government Printing Office, Washington, DC).
Eckert, W.J., Brouwer, D., and Clemence, G.M. (1951). "Coordinates of the Five Outer
Planets 1653-2060" Astro. Papers Amer. Eph. XII (U.S. Government Printing Office,
Washington, DC).
Fox, L. and Parker, LB. (1972). Chebyshev Polynomials in Numerical Analysis (Oxford
University Press, Oxford, England).
Herget, P. (1948). Computation of Orbits Published privately by author.
Hildebrand, F.B. (1987). Introduction to Numerical Analysis (Dover Press, New York).
H.M. Nautical Almanac Office (1956). Interpolation and Allied Tables (H.M. Stationery
Office, London).
Jennings, W. (1964). First Course in Numerical Methods (McMillan, New York).
Lanczos, C. (1956). Applied Analysis (Prentice-Hall, Englewood Cliff, NJ).
Press, W.H., Flannery, B.P., Teukolsky, S.A., and Vetterling, W.T. (1986). Numerical
Recipes, The Art of Scientific Computing (Cambridge University Press, Cambridge,
England) .
Scarborough, J.B. (1955). Numerical Mathematical Analysis, 3rd ed. (Johns Hopkins Uni-
versity Press, Baltimore).
Smart, W.M. (1958). Combination of Observations (Cambridge University Press, Cam-
bridge, England).
Whittaker, E.T. and Robinson, G. (1944). The Calculus of Observations, 4th ed. (Blackie,
London) .
CHAPTER
Calendars
by L.E. Doggett
12.1 INTRODUCTION
A calendar is a system of organizing units of time for the purpose of reckoning
time over extended periods. By convention, the day is the smallest calendrical unit
of time; the measurement of fractions of a day is classified as timekeeping. The
generality of this definition is due to the diversity of methods that have been used
in creating calendars. Although some calendars replicate astronomical cycles ac-
cording to fixed rules, others are based on abstract, perpetually repeating cycles of
no astronomical significance. Some calendars are regulated by astronomical obser-
vations, some carefully and redundantly enumerate every unit, and some contain
ambiguities and discontinuities. Some calendars are codified in written laws; others
are transmitted by oral tradition.
The common theme of calendar making is the desire to organize units of time
to satisfy the needs and preoccupations of society. In addition to serving practical
purposes, the process of organization provides a sense, however illusory, of under-
standing and controlling time itself. Thus calendars serve as a link between mankind
and the cosmos. It is little wonder that calendars have held a sacred status and have
served as a source of social order and cultural identity. Calendars have provided
the basis for planning agricultural, hunting, and migration cycles, for divination
and prognostication, and for maintaining cycles of religious and civil events. What-
ever their scientific sophistication, calendars must ultimately be judged as social
contracts, not as scientific treatises.
According to a recent estimate (Fraser, 1987), there are about forty calendars
used in the world today. This chapter is limited to the half-dozen principal calendars
in current use. Furthermore, the emphasis of the chapter is on function and calcu-
lation rather than on culture. The fundamental bases of the calendars are given,
575
576 EXPLANATORY SUPPLEMENT
along with brief historical summaries. Although algorithms are given for correlat-
ing these systems, close examination reveals that even the standard calendars are
subject to local variations. With the exception of the Julian calendar, this chapter
does not deal with extinct systems. Inclusion of the Julian calendar is justified by
its everyday use in historical studies.
Despite a vast literature on calendars, truly authoritative references, partic-
ularly in English, are difficult to find. Aveni (1989) surveys a broad variety of
calendrical systems, stressing their cultural contexts rather than their operational
details. Parise (1982) provides useful, though not infallible, tables for date con-
version. Fotheringham (1935) and the Encyclopedia of Religion and Ethics (1910),
in its section on "Calendars," offer basic information on historical calendars. The
sections on "Calendars" and "Chronology" in all editions of the Encyclopedia Bri-
tannica provide useful historical surveys. Ginzel (1906) remains an authoritative, if
dated, standard of calendrical scholarship. References on individual calendars are
given in the relevant sections.
12.1 1 Astronomical Bases of Calendars
The principal astronomical cycles are the day (based on the rotation of the Earth
on its axis), the year (based on the revolution of the Earth around the Sun), and the
month (based on the revolution of the Moon around the Earth). The complexity
of calendars arises because these cycles of revolution do not comprise an integral
number of days, and because astronomical cycles are neither constant nor perfectly
commensurable with each other.
The tropical year is defined as the mean interval between vernal equinoxes; it
corresponds to the cycle of the seasons. The following expression, based on the
orbital elements of Laskar (1986), is used for calculating the length of the tropical
year:
365^2421896698 -0.00000615359 7- 7.29 x 10"-10r2 + 2. 64 x 10~WT3 (12.11-1)
where T = (JD - 2451545.0) / 36525 and JD is the Julian day number (see Sec-
tion 12.7). However, the interval from a particular vernal equinox to the next may
vary from this mean by several minutes.
The synodic month, the mean interval between conjunctions of the Moon and
Sun, corresponds to the cycle of lunar phases. The following expression for the syn-
odic month is based on the lunar theory of Chapront-Touze and Chapront (1988):
29^5305888531 +0.00000021621 7- -3. 64 x 10-|0r2. (12.11-2)
Again T = (JD -2451545. 0)/36525 and JD is the Julian day number. Any particular
phase cycle may vary from the mean by up to seven hours.
12 / CALENDARS 577
In the preceding formulas, T is measured in Julian centuries of Terrestrial Dy-
namical Time (TDT), which is independent of the variable rotation of the Earth.
Thus, the lengths of the tropical year and synodic month are here defined in days
of 86400 seconds of International Atomic Time (TAI). (See Chapter 2 for further
information on time scales.)
From these formulas we see that the cycles change slowly with time. Further-
more, the formulas should not be considered to be absolute facts; they are the best
approximations possible today. Therefore, a calendar year of an integral number of
days cannot be perfectly synchronized to the tropical year. Approximate synchro-
nization of calendar months with the lunar phases requires a complex sequence of
months of 29 and 30 days. For convenience it is common to speak of a lunar year
of twelve synodic months, or 354.36707 days.
Three distinct types of calendars have resulted from this situation. A solar
calendar, of which the Gregorian calendar in its civil usage is an example, is de-
signed to maintain synchrony with the tropical year. To do so, days are intercalated
(forming leap years) to increase the average length of the calendar year. A lunar
calendar, such as the Islamic calendar, follows the lunar phase cycle without regard
for the tropical year. Thus the months of the Islamic calendar systematically shift
with respect to the months of the Gregorian calendar. The third type of calendar,
the lunisolar calendar, has a sequence of months based on the lunar phase cycle;
but every few years a whole month is intercalated to bring the calendar back in
phase with the tropical year. The Hebrew and Chinese calendars are examples of
this type of calendar.
12.12 Nonastronomical Bases of Calendars: the Week
Calendars also incorporate nonastronomical elements, such as numerical cycles,
local environmental observations, or decisions by societal authorities. In the Gre-
gorian calendar, the week and month are nonastronomical units, though the month
can be traced back to calendars that counted the phase cycle of the Moon.
The origin of the seven-day week is uncertain. As a continuously running, un-
interrupted cycle, it comes to us as part of Jewish tradition. However, Biblical
and Talmudic texts indicate a variety of conflicting calendrical practices. System-
atic observance of the Sabbath every seventh day may have developed as late as
the Babylonian Exile in the sixth century B.C., followed by a period of gradual
acceptance.
The number seven had mystical and cosmological significance throughout the
Semitic cultures. It was used in the Babylonian and Assyrian calendars, though not
as a continuous cycle. Although the seven-day week may have its origin in number
mysticism, astronomy may have contributed to the mysticism. The Sun, Moon,
and naked-eye planets comprise the seven "wandering stars." Seven is a useful, if
inexact, count of days between the four Moon phases, and four times seven is a
578 EXPLANATORY SUPPLEMENT
useful, if inexact, estimate of days from first to last visibility of the Moon in its
phase cycle.
In Jewish practice, the days of the week were designated by numbers rather
than by names, except that day 7 was known as the Sabbath. Recognition of the
week and the observance of the Sabbath as a rest day gradually spread to the
Roman world. The use of names to designate the days of the week developed in
the Roman culture of the second and first centuries B.C. These names come from
the astrological practice of naming each day after the planet that governed the
day. At that time, the Sun and Moon were included among the planets. "Saturn's
day" coincided with the Jewish Sabbath. The planetary names were gradually, if
reluctantly, accepted by the Jewish and Christian cultures (Gandz, 1949).
It is a reasonable, but ultimately unprovable, assumption that the cyclic conti-
nuity of the week was maintained without interruption by religious authorities from
its origin in Biblical times to the present day. Although ten days were dropped from
the Christian calendar in the Gregorian Reform of 1582, the cycle of weekdays was
not disturbed. Thus, in 1582, Thursday, October 4, of the Julian calendar, was
followed by Friday, October 15, of the Gregorian calendar.
Colson (1926) and Zerubavel (1985) are standard references on the week. The
latter reference also considers other week-like numerical cycles used by past and
present cultures.
12.13 Calendar Reform and Accuracy
In most societies a calendar reform is an extraordinary event. Adoption of a calendar
depends on the forcefulness with which it is introduced and on the willingness of
society to accept it. For example, the acceptance of the Gregorian calendar as a
worldwide standard spanned more than three centuries.
The legal code of the United States does not specify an official national calendar.
Use of the Gregorian calendar in the United States stems from an Act of Parliament
of the United Kingdom in 1751, which specified use of the Gregorian calendar in
England and its colonies. However, its adoption in the United Kingdom and other
countries was fraught with confusion, controversy, and even violence (Bates, 1952:
Gingerich, 1983; Hoskin, 1983). It also had a deeper cultural impact through the
disruption of traditional festivals and calendrical practices (MacNeill, 1982).
Because calendars are created to serve societal needs, the question of a calen-
dar's accuracy is usually misleading or misguided. A calendar that is based on a
fixed set of rules is accurate if the rules are consistently applied. For calendars that
attempt to replicate astronomical cycles, one can ask how accurately the cycles are
replicated. However, astronomical cycles are not absolutely constant, and they are
not known exactly (see Section 12.11). In the long term, only a purely observational
12 / CALENDARS 579
calendar maintains synchrony with astronomical phenomena. However, an obser-
vational calendar exhibits short-term uncertainty, because the natural phenomena
are complex and the observations are subject to error.
12.14 Historical Eras and Chronology
The calendars treated in this chapter, except for the Chinese calendar (see Sec-
tion 12.61), have counts of years from initial epochs. In the case of the Chinese
calendar and some calendars not included here, years are counted in cycles, with
no particular cycle specified as the first cycle. Some cultures eschew year counts al-
together but name each year after an event that characterized the year. However, a
count of years from an initial epoch is the most successful way of maintaining a con-
sistent chronology. Whether this epoch is associated with an historical or legendary
event, it must be tied to a sequence of recorded historical events.
This is illustrated by the adoption of the birth of Christ as the initial epoch
of the Christian calendar. This epoch was established by the sixth-century scholar
Dionysius Exiguus, who was compiling a table of dates of Easter. An existing table
covered the nineteen-year period denoted 228-247, where years were counted from
the beginning of the reign of the Roman emperor Diocletian. Dionysius continued
the table for a nineteen-year period, which he designated Anni Domini Nostri Jesu
Christ! 532-550. Thus, Dionysius' Anno Domini 532 is equivalent to Anno Diocle-
tiani 248. In this way a correspondence was established between the new Christian
Era and an existing system associated with historical records. What Dionysius did
not do is establish an accurate date for the birth of Christ. Although scholars gen-
erally believe that Christ was born some years before A.D. 1, the historical evidence
is too sketchy to allow a definitive dating.
Given an initial epoch, one must consider how to record preceding dates. Bede,
the eighth-century English historian, began the practice of counting years backward
from A.D. 1 (see Colgrave and Mynors, 1969). In this system, the year A.D. 1 is
preceded by the year 1 B.C., without an intervening year 0. Because of the numerical
discontinuity, this "historical" system is cumbersome for comparing ancient and
modern dates. Today, astronomers use +1 to designate A.D. 1. Then +1 is naturally
preceded by year 0, which is preceded by year —1. Since the use of negative numbers
developed slowly in Europe, this "astronomical" system of dating was delayed until
the eighteenth century, when it was introduced by the astronomer Jacques Cassini
(Cassini, 1740).
Even as use of Dionysius' Christian Era became common in ecclesiastical writ-
ings of the Middle Ages, traditional dating from regnal years continued in civil use.
In the sixteenth century, Joseph Justus Scaliger tried to resolve the patchwork of
historical eras by placing everything on a single system (Scaliger, 1583). Instead
of introducing negative year counts, he sought an initial epoch in advance of any
historical record. His numerological approach utilized three calendrical cycles: the
580 EXPLANATORY SUPPLEMENT
28-year solar cycle, the nincteen-year cycle of Golden Numbers, and the fifteen-year
induction cycle. The solar cycle is the period after which weekdays and calendar
dates repeat in the Julian calendar. The cycle of Golden Numbers is the period
after which moon phases repeat (approximately) on the same calendar dates. The
indiction cycle was a Roman tax cycle. Scaliger could therefore characterize a year
by the combination of numbers (S,G,/), where 5 runs from 1 through 28, G from
1 through 19, and / from 1 through 15. Scaliger noted that a given combination
would recur after 7980 (= 28 x 19 X 15) years. He called this a Julian Period, be-
cause it was based on the Julian calendar year. For his initial epoch Scaliger chose
the year in which S, G, and / were all equal to 1. He knew that the year 1 B.C. was
characterized by the number 9 of the solar cycle, by the Golden Number 1, and by
the number 3 of the indiction cycle, i.e., (9,1,3). He found that the combination
(1, 1, 1) occurred in 4713 B.C. or, as astronomers now say, -4712. This serves as
year 1 of Scaliger's Julian Period. It was later adopted as the initial epoch for the
Julian day numbers (see Section 12.7).
12.2 THE GREGORIAN CALENDAR
The Gregorian calendar today serves as an international standard for civil use. In
addition, it regulates the ceremonial cycle of the Roman Catholic and Protestant
churches. In fact, its original purpose was ecclesiastical. Although a variety of other
calendars are in use today, they are restricted to particular religions or cultures.
12.21 Rules for Civil Use
Years are counted from the initial epoch defined by Dionysius Exiguus (see Sec-
tion 12.14), and are divided into two classes: common years and leap years. A
common year is 365 days in length; a leap year is 366 days, with an intercalary day.
designated February 29, preceding March 1. Leap years are determined according
to the following rule:
Every year that is exactly divisible by 4 is a leap year, except for years
that are exactly divisible by 100; these centurial years are leap years
only if they are exactly divisible by 400.
As a result the year 2000 is a leap year, whereas 1900 and 2100 are not leap
years. These rules can be applied to times prior to the Gregorian reform to create
a proleptic Gregorian calendar. In this case, year 0 (1 B.C.) is considered to be
exactly divisible by 4, 100, and 400; hence it is a leap year.
The Gregorian calendar is thus based on a cycle of 400 years, which comprises
146097 days. Since 146097 is evenly divisible by 7, the Gregorian civil calendar
exactly repeats after 400 years. Dividing 146097 by 400 yields an average length
12 / CALENDARS
581
Table 12.21.1
Months of the Gregorian Calendar
1
lam inru
Jal lUal y
31
7.
July
31
2.
February
28*
8.
August
31
3.
March
31
9.
September
30
4.
April
30
10.
October
31
5.
May
31
11.
November
30
6.
June
30
12.
December
31
* In a leap year, February has 29 days.
of 365.2425 days per calendar year, which is a close approximation to the length
of the tropical year. Comparison with Equation 12.11 1 reveals that the Gregorian
calendar accumulates an error of one day in about 2500 years. Although various
adjustments to the leap-year system have been proposed, none has been instituted.
Within each year, dates are specified according to the count of days from the
beginning of the month. The order of months and number of days per month were
adopted from the Julian calendar (see Table 12.21.1 and Section 12.8).
12.22 Ecclesiastical Rules
The ecclesiastical calendars of Christian churches are based on cycles of movable
and immovable feasts. Christmas is the principal immovable feast, with its date
set at December 25. Easter is the principal movable feast, and dates of most other
movable feasts are determined with respect to Easter. However, the movable feasts
of the Advent and Epiphany seasons are Sundays reckoned from Christmas and the
Feast of the Epiphany, respectively.
In the Gregorian calendar, the date of Easter is defined to occur on the Sunday
following the ecclesiastical Full Moon that falls on or next after March 21. This
should not be confused with the popular notion that Easter is the first Sunday
after the first Full Moon following the vernal equinox. In the first place, the vernal
equinox does not necessarily occur on March 21. In addition, the ecclesiastical
Full Moon is not the astronomical Full Moon— it is based on tables that do not
take into account the full complexity of lunar motion. As a result, the date of an
ecclesiastical Full Moon may differ from that of the true Full Moon. However, the
Gregorian system of leap years and lunar tables does prevent progressive departure
of the tabulated data from the astronomical phenomena.
The ecclesiastical Full Moon is defined as the fourteenth day of a tabular luna-
tion, where day 1 corresponds to the ecclesiastical New Moon. The tables are based
on the Metonic cycle, in which 235 mean synodic months occur in 6939.688 days.
Since nineteen Gregorian years is 6939.6075 days, the dates of Moon phases in a
given year will recur on nearly the same dates nineteen years later. To prevent
582 EXPLANATORY SUPPLEMENT
the 0.08 day difference between the cycles from accumulating, the tables incorpo-
rate adjustments to synchronize the system over longer periods of time. Additional
complications arise because the tabular lunations are of 29 or 30 integral days.
The entire system comprises a period of 5700000 years of 2081882250 days, which
is equated to 70499183 lunations. After this period, the dates of Easter repeat
themselves.
The following algorithm for computing the date of Easter is based on the al-
gorithm of Oudin (1940). It is valid for any Gregorian year, Y. All variables are
integers and the remainders of all divisions are dropped. The final date is given by
At, the month, and D, the day of the month.
C=Yi 100,
N=Y— 19x<T/ 19),
K=(C- 17)/ 25,
l = C-CI 4- (C-/Q/3 + I9x 7V+ 15,
/ = /- 30x(//30),
/ = /-(//28)x(l -(//28)x(29/(/+l))x((21 -AO/11)),
J =Y+Y /4 + /+2 - C + C/4,
7 = 7-7 x {J 11),
L = I-J,
M = 3 + (L + 40)/44,
D = L + 28-31x(M/4).
Example Compute the date of Easter in +2010.
C = 2010/ 100
AT = 2010- 19 x (2010/ 19)
K = (20- 1 7) / 25
/ = 20 - 20/4 - (20 - 0)/3+ 19 x 15 + 15
/ = 309 - 30 x (309 / 30)
/=9-(9/28)x(l -{9/28)x(29/(9+l))x((21 - 15)/ 11))
= 9-0x(l -0x2x 1)
7 = 2010 + 2010/4 + 9 + 2 - 20 + 20/4
J = 2508 - 7 x (2508 / 7)
= 20,
= 15,
= 0,
= 309,
= 9,
= 9,
= 2508,
= 2,
12 / CALENDARS 583
£ = 9-2 =7,
M = 3 + (7 + 40) / 44 = 4,
D = 7+28- 31 x (4/4) =4.
Thus in +2010 Easter will occur on April 4.
1 2.23 History of the Gregorian Calendar
The Gregorian calendar resulted from a perceived need to reform the method of
calculating dates of Easter. Under the Julian calendar the dating of Easter had
become standardized, using March 21 as the date of the equinox and the Metonic
cycle as the basis for calculating lunar phases. By the thirteenth century it was
realized that the true equinox had regressed from March 21 (its supposed date
at the time of the Council of Nicea, +325) to a date earlier in the month. As a
result, Easter was drifting away from its springtime position and was losing its
relation with the Jewish Passover. Over the next four centuries, scholars debated
the "correct" time for celebrating Easter and the means of regulating this time
calendrically. The Church made intermittent attempts to solve the Easter question,
without reaching a consensus.
By the sixteenth century the equinox had shifted by ten days, and astronomical
New Moons were occurring four days before ecclesiastical New Moons. At the behest
of the Council of Trent, Pope Pius V introduced a new Breviary in 1568 and Missal
in 1570, both of which included adjustments to the lunar tables and the leap-year
system. Pope Gregory XIII, who succeeded Pope Pius in 1572, soon convened a
commission to consider reform of the calendar, since he considered his predecessor's
measures inadequate.
The recommendations of Pope Gregory's calendar commission were instituted
by the papal bull "Inter Gravissimus," signed on 1582 February 24. Ten days were
deleted from the calendar, so that 1582 October 4 was followed by 1582 October 15,
thereby causing the vernal equinox of 1583 and subsequent years to occur about
March 21. And a new table of New Moons and Full Moons was introduced for
determining the date of Easter.
Subject to the logistical problems of communication and governance in the six-
teenth century, the new calendar was promulgated through the Roman-Catholic
world. Protestant states initially rejected the calendar, but gradually accepted it
over the coming centuries. The Eastern Orthodox churches rejected the new cal-
endar and continued to use the Julian calendar with traditional lunar tables for
calculating Easter. Because the purpose of the Gregorian calendar was to regulate
the cycle of Christian holidays, its acceptance in the non-Christian world was ini-
tially not at issue. But as international communications developed, the civil rules
of the Gregorian calendar were gradually adopted around the world.
584 EXPLANATORY SUPPLEMENT
Anyone seriously interested in the Gregorian calendar should study the col-
lection of papers resulting from a conference sponsored by the Vatican to com-
memorate the four- hundredth anniversary of the Gregorian Reform (Coyne et al,
1983).
12.3 THE HEBREW CALENDAR
As it exists today, the Hebrew calendar is a lunisolar calendar that is based on
calculation rather than observation. This calendar is the official calendar of Israel
and is the liturgical calendar of the Jewish faith.
In principle the beginning of each month is determined by a tabular New Moon
(molad) that is based on an adopted mean value of the lunation cycle. To ensure that
religious festivals occur in appropriate seasons, months are intercalated according
to the Metonic cycle, in which 235 lunations occur in nineteen years.
By tradition, days of the week are designated by number, with only the seventh
day, Sabbath, having a specific name. Days are reckoned from sunset to sunset, so
that day 1 begins at sunset on Saturday and ends at sunset on Sunday. The Sabbath
begins at sunset on Friday and ends at sunset on Saturday.
12.31 Rules
Years are counted from the Era of Creation, or Era Mundi, which corresponds to
—3760 October 7 on the Julian proleptic calendar. Each year consists of twelve or
thirteen months, with months consisting of 29 or 30 days. An intercalary month
is introduced in years 3, 6, 8, 11, 14, 17, and 19 in a nineteen-year cycle of 235
lunations. The initial year of the calendar, a.m. (Anno Mundi) 1, is year 1 of the
nineteen-year cycle.
The calendar for a given year is established by determining the day of the week
of Tishri 1 (first day of Rosh Hashanah or New Year's Day) and the number of days
in the year. Years are classified according to the number of days in the year (see
Table 12.31.1).
Table 12.31.1
Classification of Years in the Hebrew Calendar
Deficient
Regular
Complete
Ordinary year
353
354
355
Leap year
383
384
385
12 / CALENDARS
585
Table 12.31 ,2
Months of the Hebrew Calendar
1.
Tishri
30
7.
Nisan
30
2.
Hestivan
29*
8.
lyar
29
3.
Kislev
30 1
9.
Sivan
30
4.
Tevet
29
10.
Tammuz
29
5.
Shevat
30
11.
Av
30
6.
Adar
29*
12.
Elul
29
* In a complete year, Heshvan has 30 days.
+ In a deficient year, Kislev has 29 days.
* In a leap year Adar I has 30 days; it is followed by Adar II with 29 days.
Table 12.31.3
Terminology of the Hebrew Calendar
Deficient (/laser) month: a month comprising 29 days.
Full {male) month: a month comprising 30 days.
Ordinary year: a year comprising 1 2 months, with a total of 353, 354, or 355 days.
Leap year: a year comprising 1 3 months, with a total of 383, 384, or 385 days.
Complete year (shelemah): a year in which the months of Heshvan and Kislev both contain
30 days.
Deficient year (haser): a year in which the months of Heshvan and Kislev both contain
29 days.
Regular year (kesidrah): a year in which Heshvan has 29 days and Kislev has 30 days.
Halakim (singular, helek): "parts" of an hour; there are 1080 halakim per hour.
Molad (plural, moladot): "birth" of the Moon, taken to mean the time of conjunction for modem
calendric purposes.
Dehiyyah (plural, dehiyyot): "postponement"; a rule delaying 1 Tishri until after the molad.
The months of Heshvan and Kislev vary in length to satisfy requirements for
the length of the year (see Table 12.31.1). In leap years, the 29-day month Adar is
designated Adar II, and is preceded by the 30-day intercalary month Adar I.
For calendrical calculations, the day begins at 6 p.m., which is designated 0
hours. Hours are divided into 1080 halakim; thus one helek is 3 1/3 seconds. (Ter-
minology is explained in Table 12.31.3.) Calendrical calculations are referred to the
meridian of Jerusalem — 2 hours 21 minutes east of Greenwich.
Rules for constructing the Hebrew calendar are given in the sections that follow.
Cohen (1981), Resnikoff (1943), and Spier (1952) provide reliable guides to the rules
of calculation.
12.311 Determining Tishri 1 The calendar year begins with the first day of Rosh
Hashanah (Tishri 1). This is determined by the day of the Tishri molad and the
four rules of postponements (dehiyyot). The dehiyyot can postpone Tishri 1 until
586
EXPLANATORY SUPPLEMENT
Table 12.311.1
Lunation Constants for Determining Tishri 1
Lunations Wee ks -Day s-H o u rs-Halakim
1 = 4-1-12-0793
12 = 50-4-08-0876
13 = 54-5-21-0589
235 = 991-2-16-0595
one or two days following the moiad. Tabular new moons (maladot) are reckoned
from the Tishri molad of the year a.m. 1, which occurred on day 2 at 5 hours, 204
halakim (i.e., 11:11:20 P.M. on Sunday, —3760 October 6, Julian proleptic calen-
dar). The adopted value of the mean lunation is 29 days, 12 hours, 793 halakim
(29.530594 days). To avoid rounding and truncation errors, calculations should be
done in halakim rather than decimals of a day, since the adopted lunation constant
is expressed exactly in halakim.
Lunation constants required in calculations are shown in Table 12.311.1. By
subtracting off the weeks, these constants give the shift in weekdays that occurs
after each cycle.
Example 1 Find the weekday and time of the Tishri jnolad of A.M. 2.
Since it is an ordinary year, A.M. 1 has twelve lunations. Therefore, the shift in weekdays
for twelve lunations can be added to the epochal molad:
Days-Hours-Ha/afa'm
epochal molad 2-05-0204
+ 12 lunations 4-08-0876
Tishri molad, a.m. 2 6-14-0000
day 6 at 14 hours 000 halakim corresponds to Thursday at 8 A.M.
The dehiyyot are as follows:
(a) If the Tishri molad falls on day 1, 4, or 6, then Tishri 1 is postponed one
day.
(b) If the Tishri m,olad occurs at or after 18 hours (i.e., noon), then Tishri 1
is postponed one day. If this causes Tishri 1 to fall on day 1, 4, or 6, then
Tishri 1 is postponed an additional day to satisfy dehiyyah (a).
(c) If the Tishri molad of an ordinary year (i.e., of twelve months) falls on day
3 at or after 9 hours, 204 halakim, then Tishri 1 is postponed two days to
day 5, thereby satisfying dehiyyah (a).
12 / CALENDARS 5Q7
(d) If the first molad following a leap year falls on day 2 at or after 15 hours
589 halakim, then Tishri 1 is postponed one day to day 3.
Example 2 Find the day of A.M. 5760 Tishri 1.
Since the initial epoch, 5759 years have passed, comprising 303 nineteen-year cycles plus
two years. Both of these years are ordinary years, The lunation constants given previously
(with whole weeks eliminated) provide the weekday displacement, which are added to the
epochal molad:
Days-Hours-Ha/aMm
19-year cycles 303 x 2-16-0595 = 2-22-1005
ordinary years 2 X 4-08-0876 = 1-17-0672
leap years 0 x 5-21-0589 = 0-00-0000
epochal molad 2-05-0204
Tishri molad, A.M. 5760 6-21-0801
Since the molad occurs on day 6, dehiyyah (a) causes Tishri 1 to be postponed to day 7.
It may be noted that dehiyyah (b) would also cause a postponement of one day, since the
molad occurs after 18 hours. Dehiyyah (a) takes precedence, however.
12.312 Reasons for the Dehiyyot Dehiyyah (a) prevents Hoshana Rabba (Tishri
21) from occurring on the Sabbath and prevents Yom Kippur (Tishri 10) from
occurring on the day before or after the Sabbath.
Dehiyyah (b) is an artifact of the ancient practice of beginning each month with
the sighting of the lunar crescent. It is assumed that if the molad (i.e., the mean
conjunction) occurs after noon, the lunar crescent cannot be sighted until after
6 P.M., which will then be on the following day. (For further information about
visibility of the lunar crescent, see Section 12.411.)
Dehiyyah (c) prevents an ordinary year from exceeding 355 days. If the Tishri
molad of an ordinary year occurs on Tuesday at or after 3:11:20 A.M., the next
Tishri molad will occur at or after noon on Saturday. According to dehiyyah (b),
Tishri 1 of the next year must be postponed to Sunday, which by dehiyyah (a)
occasions a further postponement to Monday. This results in an ordinary year of
356 days. Postponing Tishri 1 from Tuesday to Thursday produces a year of 354
days.
Dehiyyah (d) prevents a leap year from falling short of 383 days. If the Tishri
molad following a leap year is on Monday, at or after 9:32:43 1/3 A.M., the previ-
ous Tishri molad (thirteen months earlier) occurred on Tuesday at or after noon.
Therefore, by dehiyyot (b) and (a), Tishri 1 beginning the leap year was postponed
to Thursday. To prevent a leap year of 382 days, dehiyyah (d) postpones by one
day the beginning of the ordinary year.
A thorough discussion of both the functional and religious aspects of the de-
hiyyot is provided by Cohen (1981).
588 EXPLANATORY SUPPLEMENT
12.313 Determining the Length of the Year An ordinary year consists of 50 weeks
plus 3, 4, or 5 days. The number of excess days identifies the year as being deficient ,
regular, or complete, respectively. A leap year consists of 54 weeks plus 5, 6, or 7
days, which again are designated deficient, regular, or complete, respectively. The
length of a year can therefore be determined by comparing the weekday of Tishri
1 with that of the next Tishri 1.
First consider an ordinary year. The weekday shift after twelve lunations is
04-08-876. For example if a Tishri molad of an ordinary year occurs on day 2 at
0 hours 0 halakim (6 p.m. on Monday), the next Tishri molad will occur on day
6 at 8 hours 876 halakim. The first Tishri molad does not require application of
the dehiyyot, so Tishri 1 occurs on day 2. Because of dehiyyah (a), the following
Tishri 1 is delayed by one day to day 7, five weekdays after the previous Tishri 1.
Since this characterizes a complete year, the months of Heshvan and Kislev both
contain 30 days.
The weekday shift after thirteen lunations is 05-21-589. If the Tishri molad
of a leap year occurred on day 4 at 20 hours 500 halakim, the next Tishri molad
will occur on day 3 at 18 hours 9 halakim. Because of dehiyyot (b), Tishri 1 of the
leap year is postponed two days to day 6. Because of dehiyyah (c), Tishri 1 of the
following year is postponed two days to day 5. This six-day difference characterizes
a regular year, so that Heshvan has 29 days and Kislev has 30 days.
1 2.32 History of the Hebrew Calendar
The codified Hebrew calendar as we know it today is generally considered to date
from a.m. 4119 (+359), though the exact date is uncertain. At that time the pa-
triarch Hillel II, breaking with tradition, disseminated rules for calculating the
calendar. Prior to that time the calendar was regarded as a secret science of the
religious authorities. The exact details of Hillel's calendar have not come down to
us, but it is generally considered to include rules for intercalation over nineteen-year
cycles. Up to the tenth century A.D., however, there was disagreement about the
proper years for intercalation and the initial epoch for reckoning years.
Information on calendrical practices prior to Hillel is fragmentary and often
contradictory. The earliest evidence indicates a calendar based on observations of
Moon phases. Since the Bible mentions seasonal festivals, there must have been
intercalation. There was likely an evolution of conflicting calendrical practices.
The Babylonian exile, in the first half of the sixth century B.C., greatly influ-
enced the Hebrew calendar. This is visible today in the names of the months. The
Babylonian influence may also have led to the practice of intercalating leap months.
During the period of the Sanhedrin, a committee of the Sanhedrin met to
evaluate reports of sightings of the lunar crescent. If sightings were not possible,
the new month was begun 30 days after the beginning of the previous month.
Decisions on intercalation were influenced, if not determined entirely, by the state
12 / CALENDARS ggg
of vegetation and animal life. Although eight-year, nineteen-year, and longer-period
intercalation cycles may have been instituted at various times prior to Hillel II, there
is little evidence that they were employed consistently over long time spans.
For information on the seven-day week, see Section 12.12.
12.4 THE ISLAMIC CALENDAR
The Islamic calendar is a purely lunar calendar in which months correspond to the
lunar phase cycle. As a result, the cycle of twelve lunar months regresses through
the seasons over a period of about 33 years. For religious purposes, Muslims begin
the months with the first visibility of the lunar crescent after conjunction. For civil
purposes a tabulated calendar that approximates the lunar phase cycle is often
used.
The seven-day week is observed with each day beginning at sunset. Weekdays
are specified by number, with day 1 beginning at sunset on Saturday and ending
at sunset on Sunday. Day 5, which is called Jum'a, is the day for congregational
prayers. Unlike the Sabbath days of the Christians and Jews, however, Juma is not
a day of rest. Jum'a begins at sunset on Thursday and ends at sunset on Friday.
12.41 Rules
Years of twelve lunar months are reckoned from the Era of the Hijra, commem-
orating the migration of the Prophet and his followers from Mecca to Medina.
This epoch, 1 a.h. (Anno Higerae) Muharram 1, is generally taken by astronomers
(Neugebauer. 1975) to be Thursday, +622 July 15 (Julian calendar). This is called
the astronomical Hijra epoch. Chronological tables (e.g., Mayr and Spuler, 1961;
Freeman-Grenville, 1963) generally use Friday, July 16, which is designated the civil
epoch. In both cases the Islamic day begins at sunset of the previous day.
For religious purposes, each month begins in principle with the first sighting
of the lunar crescent after the New Moon. This is particularly important for es-
tablishing the beginning and end of Ramadan. Because of uncertainties due to
weather, however, a new month may be declared thirty days after the beginning of
the preceding month. Although various predictive procedures have been used for
determining first visibility (see Section 12.411), they have always had an equivocal
status. In practice, there is disagreement among countries, religious leaders, and
scientists about whether to rely on observations, which are subject to error, or to
use calculations, which may be based on poor models.
Chronologists employ a thirty-year cyclic calendar in studying Islamic history.
In this tabular calendar, there are eleven leap years in the thirty-year cycle. Odd-
numbered months have thirty days and even-numbered months have twenty-nine
590
EXPLANATORY SUPPLEMENT
Table 12.41.1
Months of Tabular Islamic Calendar
1
I .
Muharram^
7.
riajau1
30
2.
Safar
29
8.
Sha'ban
29
3.
Rabi'a I
30
9.
Ramadan*
30
4.
Rabi'a II
29
10.
Shawwal
29
5.
Jumada I
30
11.
Dhu ai-Q adah5
30
6.
Jumada II
29
12.
Dhu al-HijjahS
29*
* In a leap year, Dhu al-Hijjah has 30 days.
§ Holy months,
t Month of fasting.
days, with a thirtieth day added to the twelfth month, Dhu al-Hijjah (see Ta-
ble 12.41.1). Years 2, 5, 7, 10, 13, 16, 18, 21, 24, 26, and 29 of the cycle are des-
ignated leap years. This type of calendar is also used as a civil calendar in some
Muslim countries, though other years are sometimes used as leap years. The mean
length of the month of the thirty-year tabular calendar is about 2.9 seconds less
than the synodic period of the Moon.
12.411 Visibility of the Crescent Moon Under optimal conditions, the crescent
Moon has been sighted about 15.4 hours after the astronomical New Moon (i.e.,
conjunction) (Schaefer, 1988). Usually, however, it is not seen until it is more than
twenty-four hours old. Babylonian astronomers were the first to develop methods
for calculating first visibility, though no surviving tables are explicitly concerned
with this (Neugebauer, 1975: 533-540). The earliest known visibility tables are by
ahKhwarizmi, a nineth-century astronomer of Baghdad (King, 1987). These tables,
and many subsequent tables, were based on the Indian criterion that the Moon will
be visible if the local hour angle of the Moon at sunset is equal to or less than
78°. With the development of Islamic astronomy, more complex criteria were also
developed by Muslim astronomers.
Modern models for predicting first visibility incorporate celestial mechanics,
spherical astronomy, selenology, atmospheric physics, and ophthalmology. Bruin
(1977) was the first to prepare such a model. Ilyas (1984), recognizing that the
Islamic calendar is used around the world, introduced the concept of an "Inter-
national Lunar Dateline," west of which the Moon should be visible under good
observing conditions. Further theoretical work has been done by Schaefer (1988).
Extensive observing programs have been organized by Doggett, Seidelmann, and
Schaefer (1988) and Doggett and Schaefer (1989).
12 / CALENDARS
1 2.42 History of the Islamic Calendar
591
The form of the Islamic calendar, as a lunar calendar without intercalation, was laid
down by the Prophet in the Qur'an (Sura IX, verse 36-37) and in his sermon at the
Farewell Pilgrimage. This was a departure from the lunisolar calendar commonly
used in the Arab world, in which months were based on first sightings of the lunar
crescent, but an intercalary month was added as deemed necessary.
Caliph 'Umar I is credited with establishing the Hijra Era in A.H. 17. It is not
known how the initial date was determined. However, calculations show that the
astronomical New Moon (i.e., conjunction) occurred on +622 July 14 at 0444 UT
(assuming AT1 =1.0 hour), so that sighting of the crescent most likely occurred on
the evening of July 16.
12.5 THE INDIAN CALENDAR
As a result of a calendar reform in A.D. 1957, the National Calendar of India
is a formalized lunisolar calendar in which leap years coincide with those of the
Gregorian calendar (Calendar Reform Committee, 1957). However, the initial epoch
is the Saka Era, a traditional epoch of Indian chronology. Months are named after
the traditional Indian months and are offset from the beginning of Gregorian months
(see Table 12.51.1).
In addition to establishing a civil calendar, the Calendar Reform Committee
set guidelines for religious calendars, which require calculations of the motions of
the Sun and Moon. Tabulations of the religious holidays are prepared by the India
Meteorological Department and published annually in The Indian Astronomical
Ephemeris.
Despite the attempt to establish a unified calendar for all of India, many local
variations exist. The Gregorian calendar continues in use for administrative pur-
poses, and holidays are still determined according to regional, religious, and ethnic
traditions (Chatterjee, 1987).
12.51 Rules for Civil Use
Years are counted from the Saka Era; 1 Saka is considered to begin with the vernal
equinox of A.D. 79. The reformed Indian calendar began with Saka Era 1879, Caitra
1, which corresponds to A.D. 1957 March 22. Normal years have 365 days; leap
years have 366. In a leap year, an intercalary day is added to the end of Caitra. To
determine leap years, first add 78 to the Saka year. If this sum is evenly divisible by
4, the year is a leap year, unless the sum is a multiple of 100. In the latter case, the
year is not a leap year unless the sum is also a multiple of 400. Table 12.51.1 gives
the sequence of months and their correlation with the months of the Gregorian
calendar.
592
EXPLANATORY SUPPLEMENT
Table 12.51.1
Months of the Indian Civil Calendar
Days per Month
Correlation of Indian and Gregorian Months
1
I -
Caitra
30"
Caitra 1
March 22"
o
£.
VdlodPM let
11
O I
Vaisakha 1
April 21
3.
Jyaistha
31
Jyaistha 1
May £.2.
A
**.
A cqH ha
11
O I
Asadha 1
June 22
5.
Sravana
31
Sravana 1
July 23
6.
Bhadra
31
Bhadra 1
August 23
7.
Asvina
30
Asvina 1
September 23
8.
Kariika
30
Kartika 1
October 23
9.
Agrahayana
30
Agrahayana 1
November 22
10.
Pausa
30
Pausa 1
December 22
11.
Magha
30
Magha 1
January 21
12.
Phalguna
30
Phalguna 1
February 20
"In a leap year, Caitra has 31 days and Caitra T coincides with March 21 .
A seven-day week, exactly correlated with the western week, is kept. Days run
from midnight to midnight.
1 2.52 Principles of the Religious Calendar
Religious holidays are determined by a lunisolar calendar that is based on calcula-
tions of the actual positions of the Sun and Moon. Most holidays occur on specified
lunar dates (titkis), as is explained later; a few occur on specified solar dates. The
calendrical methods presented here are those recommended by the Calendar Re-
form Committee (1957). They serve as the basis for the calendar published in The
Indian Astronomical Ephemeris. However, many local calendar makers continue to
use traditional astronomical concepts and formulas, some of which date back 1500
years.
The Calendar Reform Committee attempted to reconcile traditional calendri-
cal practices with modern astronomical concepts. According to their proposals,
precession is accounted for and calculations of solar and lunar position are based
on accurate modern methods. All astronomical calculations are performed with
respect to a Central Station at longitude 82°30' East, latitude 23° 11' North. For
religious purposes solar days are reckoned from sunrise to sunrise.
A solar month is denned as the interval required for the Sun's apparent longi-
tude to increase by 30°, corresponding to the passage of the Sun through a zodiacal
sign (rasi). The initial month of the year, Vaisakha, begins when the true longitude
of the Sun is 23° 15' {see Table 12.52.1). Because the Earth's orbit is elliptical, the
lengths of the months vary from 29.2 to 31.2 days. The short months all occur in
the second half of the year around the time of the Earth's perihelion passage.
12 / CALENDARS
593
Table 12.52.1
Solar Months of the Indian Religious Calendar
Sun's Longitude Approximate Approximate
at Beginning Duration Gregorian
of Month of Month Date
d
1.
Vaisakha
23
15
30.9
Apr. 13
2.
Jyestha
53
15
31.3
May 14
3.
Asadha
83
15
31.5
June 14
4.
Sravana
113
15
31.4
July 16
5.
Bhadrapada
143
15
31.0
Aug. 16
6.
Asvina
173
15
30.5
Sept 16
7.
Kartika
203
15
30.0
Oct. 1 7
8.
Margasirsa
233
15
29.6
Nov. 16
9.
Pausa
263
15
29.4
Dec. 15
10.
Magha
293
15
29.5
Jan. 14
11.
Phalguna
323
15
29.9
Feb. 12
12.
Caitra
353
15
30.3
Mar. 14
Lunar months are measured from one New Moon to the next (although some
groups reckon from the Full Moon). Each lunar month is given the name of the
solar month in which the lunar month begins. Because most lunations are shorter
than a solar month, there is occasionally a solar month in which two New Moons
occur. In this case, both lunar months bear the same name, but the first month
is described with the prefix adhika, or intercalary. Such a year has thirteen lunar
months. Adhika months occur every two or three years following patterns described
by the Metonic cycle or more complex lunar phase cycles.
More rarely, a year will occur in which a short solar month will pass without
having a New Moon. In that case, the name of the solar month does not occur in
the calendar for that year. Such a decayed (ksaya) month can occur only in the
months near the Earth's perihelion passage. In compensation, a month in the first
half of the year will have had two New Moons, so the year will still have twelve
lunar months. Ksaya months are separated by as few as nineteen years and as many
as 141 years.
Lunations are divided into 30 tithis, or lunar days. Each tithi is defined by the
time required for the longitude of the Moon to increase by 12° over the longitude of
the Sun. Thus the length of a tithi may vary from about 20 hours to nearly 27 hours.
During the waxing phases, tithis are counted from 1 to 15 with the designation
Sukla. Tithis for the waning phases are designated Krsna and are again counted
from 1 to 15. Each day is assigned the number of the tithi in effect at sunrise.
Occasionally a short tithi will begin after sunrise and be completed before the next
594
EXPLANATORY SUPPLEMENT
sunrise. Similarly a long tithi may span two sunrises. In the former case, a number
is omitted from the day count. In the latter, a day number is carried over to a
second day.
1 2.53 History of the Indian Calendar
The history of calendars in India is a remarkably complex subject owing to the
continuity of Indian civilization and to the diversity of cultural influences. In the
mid-1950s, when the Calendar Reform Committee made its survey, there were about
30 calendars in use for setting religious festivals for Hindus, Buddhists, and Jain-
ists. Some of these were also used for civil dating. These calendars were based
on common principles, though they had local characteristics determined by long-
established customs and the astronomical practices of local calendar makers. In
addition, Muslims in India used the Islamic calendar, and the Indian government
used the Gregorian calendar for administrative purposes.
Early allusions to a hmisolar calendar with intercalated months are found in the
hymns from the Rig Veda, dating from the second millennium B.C. Literature from
1300 B.C. to A.D. 300, provides information of a more specific nature. A five-year
hmisolar calendar coordinated solar years with synodic and sidereal lunar months.
Indian astronomy underwent a general reform in the first few centuries A.D.,
as advances in Babylonian and Greek astronomy became known. New astronom-
ical constants and models for the motion of the Moon and Sun were adapted to
traditional calendric practices. This was conveyed in astronomical treatises of this
period known as Siddhantas, many of which have not survived. The Surya Sid-
dhanta, which originated in the fourth century but was updated over the following
centuries, influenced Indian calendrics up to and even after the calendar reform of
A..D. 1957.
Pingree (1978) provides a survey of the development of mathematical astron-
omy in India. Although he does not deal explicitly with calendrics, this material is
necessary for a full understanding of the history of India's calendars.
12.6 THE CHINESE CALENDAR
The Chinese calendar is a hmisolar calendar based on calculations of the positions
of the Sun and Moon. Months of 29 or 30 days begin on days of astronomical New
Moons, with an intercalary month being added every two or three years. Since the
calendar is based on the true positions of the Sun and Moon, the accuracy of the
calendar depends on the accuracy of the astronomical theories and calculations.
Although the Gregorian calendar is used in the Peoples' Republic of China
for administrative purposes, the traditional Chinese calendar is used for setting
traditional festivals and for timing agricultural activities in the countryside. The
Chinese calendar is also used by Chinese communities around the world.
12 / CALENDARS
595
Table 12.61.1
Chinese Sexagenary Cycle of Days and Years
Celestial Stems Earthly Branches
1 .
jia
1.
zi
rat
2.
yi
2.
chou
ox
3.
bing
3.
yin
tiger
4.
ding
4.
mao
hare
5.
wu
5.
chen
dragon
6.
ji
6.
si
snake
7.
geng
7.
wu
horse
8.
xin
8.
wei
sheep
9.
ren
9.
shen
monkey
10.
gui
10.
you
fowl
11.
xu
dog
12.
hai
pig
Year Names
1.
jia-zi
16.
ji-mao
31.
jia-wu
46.
ji-you
2.
yi-chou
17.
geng-chen
32,
yi-wei
47.
geng-xu
3.
bing-yin
18.
xin-si
33.
bing-shen
48.
xin-hai
4.
ding-rrtao
19.
ren-wu
34.
ding-you
49.
ren-zi
5.
wu-chen
20.
gui-wei
35.
wu-xu
50.
gui-chou
6.
ji-si
21.
jia-shen
36.
ji-hai
51.
jia-yin
7.
geng-wu
22.
yi-you
37.
geng-zi
52.
yi-mao
8.
xin-wei
23.
bing-xu
38.
xin-chou
53.
bing-chen
9.
ren-shen
24.
ding-hai
39.
ren-yin
54.
ding-si
10.
gui-you
25.
wu-zi
40.
gui-mao
55.
wu-wu
11.
jia-xu
26.
ji-chou
41.
jia-chen
56.
ji-wei
12.
yi-hai
27.
geng-yin
42.
yi-si
57.
geng-shen
13.
bing-zi
28.
xin-mao
43.
bing-wu
58.
xin-you
14.
ding-chou
29.
ren-chen
44.
ding-wei
59.
ren-xu
15.
wu-yin
30.
gui-si
45.
wu-shen
60.
gui-hai
12.61 Rules
There is no specific initial epoch for counting years. In historical records, dates were
specified by counts of days and years in sexagenary cycles and by counts of years
from a succession of eras established by reigning monarchs (see Section 12.62).
The sixty- year cycle consists of a set of year names that are created by pairing a
name from a list of ten Celestial Stems with a name from a list of twelve Terrestrial
Branches, following the order specified in Table 12.61.1. The Celestial Stems are
specified by Chinese characters that have no English translation; the Terrestrial
Branches are named after twelve animals. After six repetitions of the set of stems
and five repetitions of the branches, a complete cycle of pairs is completed and
596 EXPLANATORY SUPPLEMENT
a new cycle begins. The initial year (jia-zi) of the current cycle began on 1984
February 2.
Days are measured from midnight to midnight. The first day of a calendar
month is the day on which the astronomical New Moon (i.e., conjunction) is cal-
culated to occur. Since the average interval between successive New Moons is ap-
proximately 29.53 days, months are 29 or 30 days long. Months are specified by
number from 1 to 12. When an intercalary month is added, it bears the number
of the previous month, but is designated as intercalary. An ordinary year of twelve
months is 353, 354, or 355 days in length; a leap year of thirteen months is 383,
384, or 385 days long.
The conditions for adding an intercalary month are determined by the occur-
rence of the New Moon with respect to divisions of the tropical year. The tropical
year is divided into 24 solar terms, in 15° segments of solar longitude. These di-
visions are paired into twelve Sectional Terms (Jieqi) and twelve Principal Terms
(Zhongqi), as shown in Table 12.61.2. These terms are numbered and assigned
names that are seasonal or meteorological in nature. For convenience in the ex-
amples that follow, the Sectional and Principal Terms are denoted by S and P,
respectively, followed by the number. Because of the ellipticity of the Earth's orbit,
the interval between solar terms varies with the seasons.
Reference works give a variety of rules for establishing New Year's Day and
for intercalation in the lunisolar calendar. Since the calendar was originally based
on the assumption that the Sun's motion was uniform through the seasons, the
published rules are frequently inadequate to handle special cases.
The following rules (Liu and Stephenson, in press) are currently used as the
basis for calendars prepared by the Purple Mountain Observatory (1984):
(1) The first day of the month is the day on which the New Moon occurs.
(2) An ordinary year has twelve lunar months; an intercalary year has thirteen
lunar months.
(3) The Winter Solstice (term P-ll) always falls in month 11.
(4) In an intercalary year, a month in which there is no Principal Term is
the intercalary month. It is assigned the number of the preceding month,
with the further designation of intercalary. If two months of an intercalary
year contain no Principal Term, only the first such month after the Winter
Solstice is considered intercalary.
(5) Calculations are based on the meridian 120° East.
The number of the month usually corresponds to the number of the Principal
Term occurring during the month. In rare instances, however, there are months that
have two Principal Terms, with the result that a nonintercalary month will have
no Principal Term. As a result the numbers of the months will temporarily fail to
correspond to the numbers of the Principal Terms. These cases can be resolved by
strictly applying rules 2 and 3. The second example deals with such a case.
12 / CALENDARS
597
Table 12.61.2
Chinese Solar Terms
TT&rm*
Noma
Sun s
L_uiiy 1 LUUC
tjutiyoridii usiQ
Approximate
Duration
o— l
-
ucnun
Beginning of Spring
olb
r&D. 4
r— 1
Yushui
Rain Water
330
Feb. 19
29.8
o— Z.
Jingzho
Waking of Insects
345
Mar. 6
Chun fen
Spring Equinox
0
Mar. 21
30.2
b— J
Qingming
Pure Brightness
15
Apr. 5
r— o
Guyu
Grain Rain
30
Apr. 20
30.7
O — 4
Lixia
Beginning of Summer
May b
D A
r— 4
Xiaoman
urain hull
60
May 21
31.2
o c
O— D
Mangzhong
Grain in Ear
75
June 6
r— 0
xiazni
Summer Solstice
90
June 22
31.4
b- b
Xiaoshu
Slight Heat
105
July 7
r — D
Dashu
threat Heat
120
1. . t. . rin
July 23
31.4
Q "7
o— /
Licfiu
Beginning of Autumn
1 35
Aug. 8
P— 7
Chushu
Limit of Heat
150
Aug. 23
31.1
o— 8
Baifu
White Dew
165
Sept. 8
D O
r— o
Qiufen
Autumnal Equinox
180
Sept. 23
30.7
b-9
Hanfu
Gold Dew
195
Oct. 8
P-9
Shuangjiang
Descent of Frost
210
Oct. 24
30.1
S-10
Lidong
Beginning of Winter
225
Nov. 8
P-10
Xiaoxue
Slight Snow
240
Nov. 22
29.7
S-11
Daxue
Great Snow
255
Dec. 7
p-11
Dongzhi
Winter Solstice
270
Dec. 22
29.5
S-12
Xiaohan
Slight Cold
285
Jan. 6
P-12
Dahan
Great Cold
300
Jan. 20
29.5
* Terms are classified as Sectional [Jieqi) or Principal (Zhongqi). followed by the number of the term.
Example 1 (typical case): Find the Gregorian date of Chinese New Year in A.D. 1991.
The Winter Solstice (Principal Term 11) occurs on 1990 December 22, and the New Moon
which occurs on or prior to that date is on December 17. Hence we compile the following
table of phenomena:
New Moon
1990 Dec 17
month 11
P-ll
Dec 22
Winter Solstice
S-12
Jan 6
New Moon
1991 Jan 16
month 12
P-12
Jan 20
S-l
Feb 4
New Moon
Feb 15
month 1
P-l
Feb 19
S-2
Mar 6
New Moon
Mar 16
month 2
598 EXPLANATORY SUPPLEMENT
Thus Chinese New Year occurs on February 15. There are no intercalary months during
this interval.
Example 2 (extraordinary case): Find the Gregorian date of Chinese New Year in A.D.
1985. Both the Winter Solstice and a New Moon occur on 1984 December 22,
New Moon
1984 Dec 22
month 11
P-ll
Dec 22
Winter Solstice
S-12
1985 Jan 5
P-12
Jan 20
New Moon
Jan 21
month 12?
6-1
Feb 4
P-l
Feb 19
New Moon
Feb 20
month 12 Intercalary?
S— 2
Mar 5
New Moon
Mar 21
month 1?
P-2
Mar 21
The first sign of difficulty is that month 11 has both Principal Terms 11 and 12. We will
also see that the month beginning on February 20 cannot be intercalary. This is discovered
by finding the next Winter Solstice, which must occur in month 11 and by checking the
number of New Moons. In 1985, P-ll occurs on December 22, and the New Moon that
begins month 11 occurs on December 12, Working backward, we find New Moons in 1985
on November 13, October 14, September 15, August 16, July 18, June 18, May 20, April
20, March 21, February 20, and January 21. Since there are only twelve New Moons from
month 11 in 1984 to month 11 in 1985, there is no room for an intercalary month. So the
correct sequence of months is as follows:
New Moon
1984 Dec 22
month 11
P-ll
Dec 22
Winter Solstice
S-12
1985 Jan 5
P-12
Jan 20
New Moon
Jan 21
month 12
S-l
Feb 4
P-l
Feb 19
New Moon
Feb 20
month 1
S-2
Mar 5
New Moon
Mar 21
month 2
P-2
Mar 21
Chinese New Year therefore occurs on 1985 February 20.
In general, the first step in calculating the Chinese calendar is to check for
■the existence of an intercalary year. This can be done by determining the dates of
Winter Solstice and month 11 before and after the period of interest, and then by
counting the intervening New Moons.
12 / CALENDARS 5gg
Published calendrical tables are often in disagreement about the Chinese calen-
dar. Some of the tables are based on mean, or at least simplified, motions of the Sun
and Moon. Some are calculated for other meridians than 120° East. Some incor-
porate a rule that the eleventh, twelfth, and first months are never followed by an
intercalary month. This is sometimes not stated as a rule, but as a consequence of
the rapid change in the Sun's longitude when the Earth is near perihelion. However,
this statement is incorrect when the motions of the Sun and Moon are accurately
calculated.
12.62 History of the Chinese Calendar
In China the calendar was a sacred document, sponsored and promulgated by the
reigning monarch. For more than two millennia, a Bureau of Astronomy made as-
tronomical observations, calculated astronomical events such as eclipses, prepared
astrological predictions, and maintained the calendar (Needham, 1959). After all,
a successful calendar not only served practical needs, but also confirmed the con-
sonance between Heaven and the imperial court.
Analysis of surviving astronomical records inscribed on oracle bones reveals a
Chinese lunisolar calendar, with intercalation of lunar months, dating back to the
Shang dynasty of the fourteenth century B.C. Various intercalation schemes were
developed for the early calendars, including the nineteen- year and 76-year lunar
phase cycles that came to be known in the West as the Metonic cycle and Callipic
cycle.
From the earliest records, the beginning of the year occurred at a New Moon
near the winter solstice. The choice of month for beginning the civil year varied
with time and place, however. In the late second century B.C., a calendar reform
established the practice, which continues today, of requiring the winter solstice to
occur in month 11. This reform also introduced the intercalation system in which
dates of New Moons are compared with the 24 solar terms (see Section 12.61).
However, calculations were based on the mean motions resulting from the cyclic
relationships. Inequalities in the Moon's motions were incorporated as early as the
seventh century A.D. (Sivin, 1969), but the Sun's mean longitude was used for
calculating the solar terms until 1644 {Liu and Stephenson, in press).
Years were counted from a succession of eras established by reigning emperors.
Although the accession of an emperor would mark a new era, an emperor might
also declare a new era at various times within his reign. The introduction of a
new era was an attempt to reestablish a broken connection between Heaven and
Earth, as personified by the emperor. The break might be revealed by the death
of an emperor, the occurrence of a natural disaster, or the failure of astronomers
to predict a celestial event such as an eclipse. In the latter case, a new era might
mark the introduction of new astronomical or calendrical models.
600 EXPLANATORY SUPPLEMENT
Sexagenary cycles were used to count years, months, days, and fractions of a day
using the set of Celestial Stems and Terrestrial Branches described in Section 12.61.
Use of the sixty-day cycle is seen in the earliest astronomical records. By contrast
the sixty-year cycle was introduced in the first century A.D. or possibly a century
earlier {Tung, 1960; Needham, 1959). Although the day count has fallen into disuse
in everyday life, it is still tabulated in calendars. The initial year (jia-zi) of the
current year cycle began on 1984 February 2, which is the third day (bing-yin) of
the day cycle.
Western (pre-Copernican) astronomical theories were introduced to China by
Jesuit missionaries in the seventeenth century. Gradually, more modern Western
concepts became known. Following the revolution of 1911, the traditional practice
of counting years from the accession of an emperor was abolished.
12.7 JULIAN DAY NUMBERS AND JULIAN DATE
The system of Julian day numbers is a continuous count of days elapsed since the
beginning of the Julian period as defined by the sixteenth-century chronologist J.
J. Scaliger (see Section 12.14). Although Scaliger's original idea was to introduce
a count of years, nineteenth-century astronomers adapted this system to create a
count of days. John Herschel (1849) thoroughly explained the system and provided
a table of "Intervals in Days between the Commencement of the Julian Period, and
that of some other remarkable chronological and astronomical Eras."
Julian day 0 commenced at Greenwich noon on —4712 January 1, Julian pro-
leptic calendar (see Section 12.8). The Julian day number, expressed as an integer,
denotes the number of complete days elapsed since the initial epoch. The Julian
date (JD) specifies a particular instant of a day by ending the Julian day number
with a decimal fraction. For example, the Julian day number of 1990 June 25 is 244
8068, whereas the Julian date at noon is 2448068.0. The midnight that begins the
civil day is specified by subtracting 0.5 from the Julian date at noon.
A count of days (1-365) from the beginning of the year is sometimes a useful
tool for record- keeping. However, the dubious practice of calling this a Julian date
merely causes confusion with Julian day numbers.
12.8 THE JULIAN CALENDAR
The Julian calendar, introduced by Julius Caesar in —45, was a solar calendar with
months of fixed lengths. Every fourth year an intercalary day was added to maintain
synchrony between the calendar year and the tropical year. It served as a standard
for European civilization until the Gregorian Reform of +1582.
Today the principles of the Julian calendar continue to be used by chronologist s.
The Julian proleptic calendar is formed by applying the rules of the Julian calendar
12 / CALENDARS gQI
to times before Caesar's reform. This provides a simple chronological system for
correlating other calendars and serves as the basis for the Julian day numbers
(Section 12.7) .
12.81 Rules
Years are classified as normal years of 365 days and leap years of 366 days. Leap
years occur in years that are evenly divisible by 4. For this purpose, year 0 (or 1
B.C., see Section 12.14) is considered evenly divisible by 4. The year is divided into
twelve formalized months that were eventually adopted for the Gregorian calendar.
1 2.82 History of the Julian Calendar
The year —45 has been called the "year of confusion," because in that year Julius
Caesar inserted 90 days to bring the months of the Roman calendar back to their
traditional place with respect to the seasons. This was Caesar's first step in replacing
a calendar that had gone badly awry. Although the pre-Julian calendar was lunisolar
in inspiration, its months no longer followed the lunar phases and its year had lost
step with the cycle of seasons (see Michels, 1967; Bickerman, 1974). Following the
advice of Sosigenes, an Alexandrine astronomer, Caesar created a solar calendar
with twelve months of fixed lengths and a provision for an intercalary day to be
added every fourth year. As a result, the average length of the Julian calendar year
was 365.25 days. This is consistent with the length of the tropical year as it was
known at the time.
Following Caesar's death, the Roman calendrical authorities misapplied the
leap-year rule, with the result that every third, rather than every fourth, year
was intercalary. Although detailed evidence is lacking, it is generally believed that
Emperor Augustus corrected the situation by omitting intercalation from the Julian
years —8 through +4. After this the Julian calendar finally began to function as
planned.
Through the Middle Ages the use of the Julian calendar evolved and acquired
local peculiarities that continue to snare the unwary historian. There were variations
in the initial epoch for counting years, the date for beginning the year, and the
method of specifying the day of the month. Not only did these vary with time and
place, but also with purpose. Different conventions were sometimes used for dating
ecclesiastical records, fiscal transactions, and personal correspondence.
Caesar designated January 1 as the beginning of the year. However, other con-
ventions flourished at different times and places. The most popular alternatives
were March 1, March 25, and December 25. This continues to cause problems for
historians, since, for example, +998 February 28 as recorded in a city that began
its year on March 1, would be the same day as +999 February 28 of a city that
began the year on January 1.
602
EXPLANATORY SUPPLEMENT
Table 12.82.1
Roman Dating in the Julian Calendar
January
March
April
August
February
May
June
December
July
September
October
November
1
Kalends
Kalends
Kalends
Kalends
2
IV
IV
VI
IV
3
III
III
V
III
4
II
II
IV
II
5
Nones
Nones
III
Nones
6
VIII
VIII
II
VIII
7
VII
VII
Nones
VII
8
VI
VI
VIII
VI
9
V
V
VII
V
10
IV
IV
VI
IV
11
III
III
V
III
12
II
II
IV
II
13
Ides
Ides
ill
Ides
14
XIX
XVI
II
XVIII
15
XVIII
XV
Ides
XVII
16
XVII
XIV
XVII
XVI
17
XVI
XIII
XVI
XV
18
XV
XII
XV
XIV
19
XIV
XI
XIV
XIII
20
XIII
X
XIII
XII
21
XII
IX
XII
XI
22
XI
VIII
XI
X
23
X
VII
X
IX
24
IX
VI
IX
VIII
25
VIII
V
VIII
VII
26
VII
IV
VII
VI
27
VI
III
VI
V
28
V
II
V
IV
29
IV
IV
III
30
III
III
II
31
II
II
Days within the month were originally counted from designated division points
within the month: Kalends, Nones, and Ides. The Kalends is the first day of the
month. The Ides is the thirteenth of the month, except in March, May, July, and
October, when it is the fifteenth day. The Nones is always eight days before the Ides
(see Table 12.82.1). Dates falling between these division points are designated by
counting inclusively backward from the upcoming division point. Intercalation was
12 / CALENDARS gQg
performed by repeating the day VI Kalends March, i.e., inserting a day between VI
Kalends March (February 24) and VII Kalends March (February 23).
By the eleventh century, consecutive counting of days from the beginning of
the month came into use. Local variations continued, however, including counts of
days from dates that commemorated local saints. The inauguration and spread of
the Gregorian calendar resulted in the adoption of a uniform standard for recording
dates.
Cappelli (1930), Grotefend and Grotefend (1941), and Cheney (1945) offer guid-
ance through the maze of medieval dating.
12.9 CALENDAR CONVERSION ALGORITHMS
Conversion between calendar systems is perhaps most easily and generally accom-
plished by use of Julian day numbers. For calendars based on fixed, consistent
rules, calendar dates can be converted to Julian day numbers and Julian day num-
bers to calendar dates. Tables for this purpose were prepared by Harvey (1983). A
generalized set of algorithms was prepared by Hatcher (1985). Parisot (1986) has
contributed parameters for Hatcher's algorithms for a few additional calendars.
The algorithms given in this section do not conform to Hatcher's general system
but were inspired by the algorithms created by Fliegel and Van Flandern (1968)
for conversion between Gregorian calendar dates and Julian day numbers. Other,
possibly more efficient, formulations are possible. Integer arithmetic is exclusively
used in this algorithms; i.e., the remainders of all divisions are dropped.
Julian day numbers run from noon to noon. Thus a calculated Julian day
number pertains to the noon occurring in the corresponding calendar date. The
notation is:
JD = Julian day number
Y = calendar year
M = month
D = day of month
All variables are integer. Since Julian day numbers for current times consist of seven
digits, a word size of at least four bytes is required to use these algorithms.
12.91 Converting Day of the Week
A problem common to several of the calendars is computing the day of the week.
The following formula gives the day of the week for the Julian day number that
pertains at noon. Given: JD. Compute: Day of week, /, where / runs from 1 though
7, with 1 being Sunday.
/ = JD - 7 x ((JD+ l)/7) + 2
(12.91-1)
604
EXPLANATORY SUPPLEMENT
Example 1 Compute the day of the week of JD 2451545.
/= 245 1 545 - 7 x ((245 1 545 + 1 ) / 7) + 2
= 2451545 -7 x 350220 + 2
= 7
Therefore JD 2451545 is Saturday.
12.92 Converting between Gregorian Calendar Date and Julian Day Number
These algorithms by Fliegel and Van Flandern (1968) are valid for all Gregorian
calendar dates corresponding to JD > 0, i.e., dates after —4713 November 23. Given:
Y, M, D. Compute: JD.
JD = (1461 x <Y + 4800 + (M- 14)/ 12))/4 + (367 x (M - 2 - 12 x ((M - 14)/ 12)))/ 12
- (3 x 07 + 4900 + CM - 14) / 12) / 100)) I4 + D- 32075 (12.92-1)
Given: JD. Compute: Y, M, D,
JD + 68569
(4 xL)l 146097
L - (146097 x N + 3)/ 4
(4000x(L+l))/ 1461001
£-(1461 x/)/4 + 31
(80 x L) 1 2447
L - (2447 x J) I 80
J I 11
7 + 2- 12xL
100x(A/-49) + /+L
(12.92-2)
L =
N =
L =
1 =
L =
7 =
D =
L =
M =
Y =
1 2.93 Converting Between Islamic Tabular Calendar Date
and Julian Day Number
While the religious calendar is based on first sighting of the lunar crescent after
New Moon, various tabular calendars have been used. In the system most often
used by historians, years 2, 5, 7, 10, 13, 16, 18, 21, 24, 26, and 29 of a 30-year
cycle are designated leap years. The following algorithms are valid for values of
Y > 1 and JD > JD0, where Y is the year of the Hijra era and JD0 is the Julian
day number of the Hijra epoch. Two different initial epochs have been used for
12 / CALENDARS QQ5
the Islamic calendar (see Section 12.41). If the astronomical epoch is preferred, in
which A.H. 1 Muharram 1 corresponds to a.d. 622 July 15, then JD0 = 1948439
should be used in the equations below; if the civil epoch is preferred, in which A.H.
1 Muharram 1 corresponds to +622 July 16, use JD0 = 1948440. Given: Y, M, D.
Compute: JD.
JD = (1 1 x y + 3) / 30 + 354 xY+ 30 xM-(M- l)/2+D + JD0 - 385 (12.93-1)
Given: JD. Compute: Y, M, D.
L = JD - JD0 + 10632
N ~(L— 1)/ 10631
L = L 10631 xN+ 354
7 sa ((10985 — L) / 53 16) x ((50 xL) I 17719) + (L / 5670) x ((43 xL)l 15238)
L = L- ((30-/)/ 15)x ((17719x7)/ 50) - (7/ 16) x ((15238 x 7) / 43) + 29
M = (24 x L) 1 709
D - L - (709 x M) I 24
T = 30xA/ + /- 30 (12.93-2)
12.94 Converting between Indian Civil Calendar and Julian Day Number
These algorithms are for the civil calendar recommended by the Calendar Reform
Committee of India. Years are counted from the Saka Era. The algorithms are valid
for all values of Y > 1 , where Y is the year reckoned from the Saka Era, and JD >
1749995. Given: 7, M, D. Compute: JD.
JD = 365 xy + (K+78- }/M)/4 + 3\ xM-(M + 9)/ll -(M !l)x{M -1)
-(3x((F + 78- 1 !M)1 100+ 1))/ 4 +D+ 1749579 (12.94-1)
Given: JD. Compute: Y, M, D.
L = JD + 68518
N = (4 x L) 1 146097
L = L (146097 x/V+3)/4
/= (4000 x (L+ 1))/ 1461001
L = 7-(1461 x/)/4+ 1
J ={{L— l)/31)x(l — Li 185) + (Z./185)x((L- 156) / 30 + 5) - L 1 366
D = L - 31 x 7 + ((7 + 2) / 8) x (7 - 5)
606 EXPLANATORY SUPPLEMENT
1 = 1/ 11
M = J + 2- \2xL
Y = 100x(Af-49)+L + /-78 (12.94-2)
12.95 Converting between Julian Calendar Date and Julian Day Number
These algorithms are valid for all values of Y > —4712, i.e., for all dates with JD >
0. The formula for computing JD from Y, M, D was constructed by Fliegel (1990) as
an entry in "The Great Julian Day Contest," held at the Jet Propulsion Laboratory
in 1970. Given: Y, M, D. Compute: JD.
JD = 367 xY — (7 X(Y+ 500 1 + (Af — 9) / 7)) / 4 + (275 x M) 1 9 + D + 1 729777
Given: JD. Compute: Y, M, D.
J =
JD + 1402
K =
(J- 1)/ 1461
L =
J - 1461 xK
N =
(L- 1)/ 365 — LI 1461
1 =
£,-365 XA/ + 30
7 =
(80 X/)/ 2447
D =
/ - (2447 x J) 1 80
1 =
J 1 11
M =
J + 2- 12x1
y =
4x£ + A/ + /-4716
12.10 REFERENCES
REFERENCES
Aveni, A.F. (1989). Empires of Time New York.
Bates, R.S. (1952). "Give Us Back Our Fortnight" Sky and Telescope 11, 267-268.
Bickerman, E.J. (1974). Chronology of the Ancient World (Ithaca).
Bruin, F. (1977). "The First Visibility of the Lunar Crescent" Vistas in Astronomy 21,
331-358.
Calendar Reform Committee (1955). Report of the Calendar Reform Committee (New
Delhi).
Cappelli, A. (1930). Cronologia, Cronografia e Calendario perpetuo Milan.
Cassini, J. (1740). Tables astronomiques du Soleil, de la Lune..., Paris.
Chapront-Touze, M. and Chapront, J. (1988). "ELP 2000-85: a Semi-Analytical Lunar
Ephemeris Adequate for Historical Times" Astron. Astrophys. 190, 342-352.
12 / CALENDARS
607
Chatterjee, S.K. (1987). 'Indian Calendars" in History of Oriental Astronomy G. Swarup
et al, eds. Cambridge, pp. 91-95
Cheney, C.R. (1945). A Handbook of Dates for Students of English History London.
Cohen, J.R. (1981). Mishnah Tractate Rosh Hashanah New York.
Colgrave, B. and Mynors, R.A.B. (1969). Bede's Ecclesiastical History of the English Peo-
ple Oxford.
Colson, F.H. (1926). The Week Cambridge.
Coyne, G.V., Hoskin, M.A., and Pedersen, O., eds. (1983). Gregorian Reform of the Cal-
endar Vatican City.
Doggett, L.E. and B.E. Schaefer (1989). "Results of the July Moonwatch" Sky & Telescope
77, 373-375.
Doggett, L.E., Seidelmann, P.K., and Schaefer, B.E. (1988). "Moonwatch — July 14, 1988"
Sky & Telescope 76, 34-35.
Encyclopedia of Religion and Ethics (1910).
Fliegel, H.F. and Van Flandern, T.C. (1968). "A Machine Algorithm for Processing Cal-
endar Dates" Communications of the Association of Computing Machines 11, 657.
Fliegel, H.F. (1990). Personal communication.
Fotheringham, J.K. (1934). "The Calendar" in The Nautical Almanac 1935 London,
Fraser, J.T. (1987). Time; the Familiar Stranger Amherst.
Freeman-Grenville, G.S.P. (1963). The Muslim and Christian Calendars London.
Gandz, S. (1949). "The Origin of the Planetary Week or the Planetary Week in Hebrew
Literature" Proceedings of the American Academy for Jewish Research 18, 213-254.
Gingerich, O. (1983). "The Civil Reception of the Gregorian Calendar" in Gregorian Re-
form of the Calendar Coyne, G.V., Hoskin, M.A., and Pedersen, O., eds., Vatican
City, pp. 265-279.
Ginzel, F.K. (1906, 1911). Handbuch der Mathematischen und Technischen Chronologie
Leipzig.
Grotefend, H. and Grotefend, O. (1941). Taschenbuch der Zeitrechnung des deutschen
Mittelalters und der Neuzeit Hannover.
Harvey, O.L. (1983). Calendar Conversions by Way of the Julian Day Number Philadel-
phia.
Hatcher, D.A. (1985). "Generalized Equations for Julian Day Numbers and Calendar
Dates" Quarterly Journal of the Royal Astronomical Society 26, 151-155.
Herschel, J.F.W. (1849). Outlines of Astronomy pp. 633-637.
Hoskin, M. (1983). "The Reception of the Calendar by Other Churches" in Gregorian
Reform of the Calendar Coyne, G.V., Hoskin, M.A., and Pedersen, O., eds., Vatican
City, pp. 255-264.
Ilyas, M. (1984). A Modem Guide to Astronomical Calculations of Islamic Calendar,
Times and Qibla Kuala Lumpur.
King, D.A. (1987). "Some Early Islamic Tables for Determining Lunar Crescent Visibility"
From Deferent to Equant: A Volume of Studies in the History of Science in the Ancient
and Medieval Near East in Honor of E.S. Kennedy King, D.A. and Saliba, G., eds.
Annals of the New York Academy of Sciences, 500, New York.
Laskar, J. (1986). "Secular Terms of Classical Planetary Theories Using the Results of
General Relativity" Astron. Astrophys. 157, 59-70.
Liu Baolin and Stephenson, F.R, "The Chinese Calendar and Its Operational Rules" In
press.
MacNeill, M. (1982). The Festival of Lughnasa Dublin.
Mayr, J. and Spuler, B. (1961). Wiistenfeld-Maler'sche Vergleichungs-Tabellen, Wiesbaden.
608
EXPLANATORY SUPPLEMENT
Michels, A.K. (1967). The Calendar of the Roman Republic Westport.
Neugebauer, O. (1975). A History of Ancient Mathematical Astronomy Part HI, New York.
Parise, F. (1982). The Book of Calendars New York.
Parisot, J. P. (1986). "Additif to the Paper of D.A. Hatcher: 'Generalized Equations for Ju-
lian Day Numbers and Calendar Dates'" Quarterly Journal of the Royal Astronomical
Society 27, 206-507.
Pingree, D. (1978). "History of Mathematical Astronomy in India" Dictionary of Scientific
Biography XV, New York, pp. 533-633.
Purple Mountain Observatory (1984). The Newly Compiled Perpetual Chinese Calendar
(1840-2050) (Popular Science Publishing House, Beijing).
Oudin, J.-M. (1940). "Etude sur la Date de Paques" Bull. Astronomique (2) 12, 391-410.
Resnikoff, L.A. (1943). "Jewish Calendar Calculations" Scripta Mathematica 9, 191-195,
274-277.
Scaliger, J.J. (1583). De emendatione temporum Paris.
Schaefer, B.E. (1988). "Visibility of the Lunar Crescent" Q. Jour. R.A.S. 29, 511-523.
Sivin, N. (1969). Cosmos and Computation in Early Chinese Mathematical Astronom-y
Leiden.
Spier, A. (1952). The Comprehensive Hebrew Calendar New York.
Tung Tso-Pin (1960). Chronological Tables of Chinese History Hong Kong.
Welch, W.C. (1957). Chinese-American Calendar for the 40th through the 89th Year of
the Chinese Republic Washington.
Zerubavel, E. (1985). The Seven Day Circle New York.
CHAPTER
Historical Information
compiled by J. Weeks
13.1 HISTORY OF THE ALMANACS
From earliest times humans have been interested in the Sun, Moon, planets, and
stars, and in determining their positions. In particular, they are interested in pre-
dicting the positions of solar-system bodies and interesting phenomena, such as
eclipses, planetary groupings and directions, and times of rising and setting. The
results of these interests were numerous mathematical approaches to the calcula-
tions, observations of limited accuracy, and attempts to predict the positions of the
"wandering" objects.
Eventually national publications of epheinerides appeared in various countries:
in France, the Connaissance des Temps (1679); in England, The Nautical Alma-
nac and Astronomical Ephemeris (1767); in Germany, the Berliner Astronomisches
Jahrbuch (1776); in Spain, the Ephemerides Astronomicas (1791); in the United
States, The American Ephemeris and Nautical Almanac (1855); in the Soviet Union,
the Astronomical Yearbook of the USSR (1923); in Japan, the Japanese Ephemeris
(1943); and in India, the Indian Ephem.eris and Nautical Almanac (1958). Initially,
these publications were based on various theories of the motions of the bodies, differ-
ent astronomical constants, different geographical coordinate systems, and different
timescales.
International cooperation led to the use of standard reference systems, systems
of constants, and common timescales. The brief histories that follow are concerned
with the major changes of form and content.
^^0V> t/ioyiovv>i'c e><
13.11 The American Ephemeris
"The Commissioners of Longitude, in pursuance of the Powers vested in
them by a late Act of Parliament, present the Publick with the NAUTI-
CAL ALMANAC and ASTRONOMICAL EPHEMERIS for the Year 1767,
609
610 EXPLANATORY SUPPLEMENT
to be continued annually; a Work which must greatly contribute to the Im-
provement of Astronomy, Geography, and Navigation. This EPHEMERIS
contains every Thing essential to general Use that is to be found in any
Ephemeris hitherto published, with many other useful and interesting Par-
ticulars never yet offered to the Publick in any Work of this Kind. The
Tables of the Moon had been brought by the late Professor MAYER of Got-
tingen to a sufficient Exactness to determine the Longitude at Sea, within a
Degree, as appeared by the Trials of several Persons who made Use of them.
The Difficulty and Length of the necessary Calculations seemed the only
Obstacles to hinder them from becoming of general Use: To remove which
this EPHEMERIS was made; the Mariner being hereby relieved from the
Necessity of calculating the Moon's Place from the Tables, and afterwards
computing the Distance to Seconds by Logarithms, which are the principal
and only very delicate Part of the Calculus; so that the finding the Longi-
tude by the Help of the EPHEMERIS is now in a Manner reduced to the
Computation of the Time, an Operation..."
"All the Calculations of the EPHEMERIS relating to the Sun and
Moon were made from Mr. MAYER'S last manuscript Tables, received by
the Board of Longitude after his Decease, which have been printed under my
Inspection, and will be published shortly. The Calculations of the Planets
were made from Dr. HALLEY'S Tables; and those of..."
These extracts from the preface to the first edition of The Nautical Almanac
and American Ephemeris (1767) were written by Nevil Maskelyne, then Astronomer
Royal. The main incentive for, and the main emphasis of, the publication was
the determination of longitude at sea using the method of lunar distances. The
ephemerides were all given in terms of apparent solar time, for the reasons given in
the book's Explanation:
"It may be proper first to premise, that all the Calculations are made
according to apparent Time by the Meridian of the Royal Observatory at
Greenwich."
"What has been shown concerning the Equation of Time chiefly re-
spects the Astronomer, the Mariner having little to do with it in computing
his Longitude from the Moon's Distances from the Sun and Stars observed
at Sea with the Help of the Ephemeris, all the Calculations thereof being
adapted to apparent Time, the same which he will obtain by the Altitudes
of the Sun or Stars in the Manner hereafter prescribed."
"But if Watches made upon Mr. John Harrison's or other equivalent
Principles should be brought into Use at Sea, the apparent Time, deduced
from an Altitude of the Sun must be corrected by the Equation of Time,
and the mean Time found compared with that shewn by the Watch, the
13 / HISTORICAL INFORMATION g-| -j
Difference will be the Longitude in Time from the Meridian by which the
Watch was set; as near as the Going of the Watch can be depended upon."
Apart from many changes in the sources of the data in the tables from which the
Moon's position was calculated, the main pages of the Almanac remained essentially
unchanged until 1834. For that year, to quote from the preface:
"The NAUTICAL ALMANAC and ASTRONOMICAL EPHEMERIS for
the Year 1834, has been constructs ed in strict conformity with the rec-
ommendations of the ASTRONOMICAL SOCIETY of LONDON, as con-
tained in their Report. . .; and will, it is believed, be found to contain almost
every aid that the Navigator and Astronomer can require."
The changes were both fundamental and substantial, and almost doubled the
size of the book. The most fundamental change was to replace apparent time as the
argument of the ephemerides with mean time. In the words of the report:
"The attention of the Committee was, in the first instance, directed to a
subject of general importance, as affecting almost all the results in the
Nautical Almanac; viz., whether the quantities therein inserted should in
future be given for apparent time (as heretofore), or for mean solar time.
Considering that the latter is the most convenient, not only for every pur-
pose of Astronomy, but also (from the best information they have been able
to obtain) for all the purposes of Navigation; at the same time that it is
less laborious to the computer, and has already been introduced with good
effect into the national Ephemerides of Coimbra and Berlin, the Committee
recommend the abolition of the use of apparent time in all the computations
of the Nautical Almanac; excepting..."
The recognition of the importance of practical astronomy also played a role, as
noted in the report:
"And here perhaps it may be proper to remark, that, although in these
discussions the Committee have constantly kept in view the principal object
for which the Nautical Almanac was originally formed, viz, the promotion
and advancement of nautical astronomy, they have not been unmindful
that, by a very slight extension of the computations, and by a few additional
articles (of no great expense or labour) , the work might be rendered equally
useful for all the purposes of practical astronomy."
The requirements of the navigator were by no means overlooked; in particular,
the number and presentation of "lunar distances," including distances from the
g-|2 EXPLANATORY SUPPLEMENT
planets, was greatly improved. However, a completely new explanation was written
in which little reference was made to the use of the ephemerides for navigation.
Tables of refraction were excluded and no example was given of clearing an observed
lunar distance for the effects of semidiameter, parallax, and refraction.
Apart from the omission of lunar distances in 1907, the first part of the almanac,
containing the ephemerides of the Sun and Moon, remained unchanged in form
until 1931. From time to time, the data was based on different data and tables,
and occasionally new material was added, such as ephemerides of the Moon and
planets at transit on the Greenwich meridian and the apparent places of many
more stars. Later, ephemerides for physical observations were included; and in 1929,
anticipating the redesign in 1931, ephemerides of the Sun referred to the standard
equinox of 1950.0 were given for the years 1928 and 1929.
Much of the added matter was of no interest to the practical navigator. In
1896, "Part I (containing such data as are more particular required for navigational
purposes)" was "also published separately for the convenience of sailors." This
consisted of a straight reprint of the monthly pages comprising the first part of
the almanac, with selections from the other data and a few specially prepared
pages. In the Preface to the 1914 edition, it was announced briefly that "Part I has
been remodeled for the convenience of sailors" ; thus was introduced The Nautical
Almanac, Abridged for the Use of Seamen, which was specially designed for its
purpose. This almanac was redesigned in 1929 and in 1952, when it was renamed
The Abridged Nautical Almanac; it was rearranged in a different form in 1958. Since
1960, it takes on the appropriate portion of the original title, namely The Nautical
Almanac.
Prior to the revision in 1931, a fundamental change in the measure of mean
solar time had taken place, which required changes in the almanac. Before 1925, the
astronomical day was considered to start at noon, and the principal ephemerides
had been given for 0h (i.e. noon) on each day. As of 1925 January 1 the tabular
day was brought into coincidence with the civil day and was considered to start at
midnight; the ephemerides were still given for 0h, now indicating midnight.
The revision of 1931 was much more than a rearrangement of the same data
into a different form. The changes of page size, of presentation, of provision for
interpolation, and of content were less important than the complete break with the
century old layout designed primarily for navigation. The new form was designed for
the astronomer without considering the requirements of navigation. Its arrangement
remained basically unchanged, though there were frequent changes in content of the
less fundamental matter.
Major changes were introduced in the 1960 edition when the almanac was
unified with The American Ephemeris, including the use of ephemeris time, in-
stead of Universal Time (mean solar time on the meridian of Greenwich), as the
argument for the fundamental ephemerides. This change further emphasized the
unsuitability of the volume for navigation, and led to the adoption of its new title:
13 / HISTORICAL INFORMATION _„ „
Dl 3
The Astronomical Ephemeris. The changes are fully described in the preface to
the edition for 1960. (Some additional notes on the history of The Astronomical
Ephemeris are given on pages ix-xviii of the volume for 1967, the two-hundredth
anniversary edition.)
13.12 The American Ephemeris and Nautical Almanac
During the first half of the nineteenth century, The Nautical Almanac remained in
general use on American ships and among astronomers and surveyors in the United
States. However, with the continued development of the country, and its growth
as a maritime nation, an increasing need for a national almanac was felt. This
need eventually led to a bill in Congress, approved in 1849, to establish a Nautical
Almanac Office in the Navy Department. The Office was set up, during the latter
part of 1849 in Cambridge, Massachusetts, where library and printing facilities were
available. The first volume of The American Ephemeris and Nautical Almanac was
for the year 1855, and was published in 1852. The office was moved to Washington
in 1866, and then was moved again to the Naval Observatory in 1893.
For the years 1855-1915 the volume was divided into two parts. Then, begin-
ning with 1882, it was divided into three parts. The first part during this entire
period was an ephemeris for the use of navigators that was also reprinted sepa-
rately, with the inclusion of a few pages from the remainder of the volume, as The
American Nautical Almanac. It comprised twelve monthly sections, for the meridian
of Greenwich, each containing ephemerides of the Sun, Moon, and lunar distances
for the month. Following the monthly sections were ephemerides of Venus, Mars,
Jupiter, and Saturn for the year, and, beginning with 1882, of Mercury, Uranus,
and Neptune.
The second part of the volume contained ephemerides of the Sun, Moon, plan-
ets, and principal stars, for meridian transit at Washington; and data on eclipses,
occultations, and a few other phenomena. In 1882, the sections on phenomena were
formally grouped as a third part with the title "Phenomena." The explanatory
sections and a few miscellaneous tables completed the volume.
During the period 1855 1915, few changes were made in the form or content.
The nautical part remained virtually unaltered; lunar distances were omitted, be-
ginning with 1912, but a page explaining how to calculate them was included. The
principal revisions in the other parts of the volume were in 1882 and 1912 1913.
The rearrangement of the 1882 volume was accompanied by some additions and
omissions. The principal omission was the ephemeris of Moon-culminating stars for
determining longitude. The principal additions were: the physical ephemerides of
Mercury and Venus for the reduction of meridian and photometric observations;
daily diagrams of the configurations of the four great satellites of Jupiter; and
ephemerides for the identification of the satellites of Mars, Saturn, Uranus, and
Neptune. In the volume for 1912, the ephemerides of the satellites were extended
g1 4 EXPLANATORY SUPPLEMENT
to include tables for determining the approximate position angle and apparent dis-
tance; in 1913, physical ephemerides were added for the Sun, Moon, Mars, and
Jupiter.
Although the volume for 1916 (the first to be issued under the international
agreements resulting from the Paris Conference of 1911) contained few content
changes, extensive revisions were made in the form and arrangement that had
been retained essentially unchanged since 1882. The arrangement of the Green-
wich ephemerides of the Sun and Moon by monthly sections was discontinued, and
replaced by annual ephemerides. At the same time, The American Nautical Alma-
nac ceased to be a reprint of part of The American Ephemeris, but instead was a
separately prepared volume especially designed for the navigator.
In 1925, the astronomical reckoning of time from 0h at noon was replaced by
the civil reckoning from midnight.
During the interval from 1916 until the fundamental revisions in 1960 when The
American Epkemeris was unified with The Astronomical Epkemeris, the revisions
of form and content were mostly only in details, although a few major changes
were incorporated. In the volumes for 1934-1937 a number of further subdivisions
and rearrangements of the contents were made. In 1937, the volume was formally
divided into seven parts; the part constituting the ephemeris for Washington was
reduced to only ephemerides of the Sun, Moon, and planets for meridian transit at
Washington— all the other material was transferred to other parts and referred to
the Greenwich meridian.
The usefulness of the Washington-transit ephemerides was limited to observers
on the Washington meridian, therefore the publication of this part was discontinued
beginning with the 1951 volume. The other principal changes in content during
1916-1959 were the following: in 1919, tables of the rising and setting of the Sun
and the Moon were added; in 1941, the number of stars for which apparent places
were given was decreased to 212 (after reaching a high of 887) when Apparent Places
of Fundamental Stars was first published; in 1957, apparent places were omitted
entirely, but precise mean places of 1551 stars, which had been given beginning in
1951, were continued. The departure of the Moon from gravitational theory due
to the variations in the rotation of the Earth made it necessary to successively
extend the elements and predictions of occultations to more and fainter stars, and
additional standard stations. An ephemeris of Pluto was added to the planetary
ephemerides in 1950; and ephemerides of Ceres, Pallas, Juno, and Vesta in 1952.
1 3.13 The Cooperative British and American Almanacs (1960 to Present)
Beginning with the editions of 1960, The Astronomical Ephemeris and The Amer-
ican Ephemeris and Nautical Almanac were joint publications. The contents were
13 / HISTORICAL INFORMATION g15
prepared cooperatively and with the exception of a few pages prior to the tabular
data, the volumes were identical. The title The Astronomical Ephemeris replaced
the title of The Nautical Almanac and the Astronomical Ephemeris. The title The
Nautical Almanac was used both in the United Kingdom and the United States
for the unified edition of the almanac for surface navigation, which had previously
been titled The Abridged Nautical Almanac, and The American Nautical Almanac,
respectively. In accordance with the resolution of the IAU, the designation ephem-
eris time was adopted for the argument of the fundamental ephemeris of the Sun,
Moon, and planets.
From 1960 onward, The Apparent Places of Fundamental Stars was published
by the Astronomisches Rechen Institute in Heidelberg, Germany, under the auspices
of the International Astronomical Union.
The IAU System of Constants adopted by the IAU at its Twelfth General
Assembly in Hamburg, Germany in 1964, were introduced with the edition for
1968. The supplement to the 1968 edition was bound in the back of the volume
and provided the details concerning the changes in the constants. An appendix
was bound in the front for the years 1968 through 1971 and gave formulas and
corrections for conversion to the IAU System of Astronomical Constants. From
1972 through 1980, the editions were published with only minor physical changes
in the arrangements of the publication.
Beginning with the edition for 1981, the title The Astronomical Almanac re-
placed the title The American Ephemeris and Nautical Almanac and the title The
Astronomical Ephemeris. This edition of The Astronomical Almanac was published
jointly by the U.S. Government Printing Office and Her Majesty's Stationery Of-
fice. The physical content of The Astronomical Almanac changed in many respects.
However, the bases for the ephemerides were unchanged, except that the funda-
mental heliocentric ephemerides of the Earth, Mercury, and Mars were computed
directly from Newcomb's theories instead of from Newcomb's Tables, and for the
Earth, the IAU 1964 System of Astronomical Constants was used.
In 1984, the bases for the ephemerides were changed. New fundamental ephem-
erides of the planets and the Moon were prepared at the Jet Propulsion Laboratory
and titled DE200/LE200. These ephemerides were in general accord with the recom-
mendations of the IAU 1976 System of Astronomical Constants, except for minor
modifications necessary to better fit the observations. The dynamical timescales
and the standard of reference of J2000.0 were introduced. The day numbers and
star positions were referred to the equinox and the equator of the middle of the
year instead of the beginning of the year. A supplement to The Astronomical Al-
manac for 1984 was bound in the back of the volume, and gave the details of the
resolutions adopted by the International Astronomical Union and documentation
concerning the ephemerides.
g1 g EXPLANATORY SUPPLEMENT
13.2 HISTORY OF INTERNATIONAL COOPERATION
13.21 Cooperation Prior to the lAU
Formal international cooperation began in October 1884 at the International Merid-
ian conference held in Washington. The resolutions of that conference included:
"the adoption of the meridian passing through the center of the transit
instrument at the Observatory of Greenwich as the initial meridian for
longitude."
"That from this meridian longitude shall be counted in two directions
up to 180 degrees, east longitude being plus and west longitude minus."
"the adoption of a universal day for all purposes for which it may be
found convenient"
"That this universal day is to be a mean solar day; is to begin for all the
world at the moment of mean midnight of the initial meridian, coinciding
with the beginning of the civil day and date of that meridian; and is to be
counted from zero up to twenty-four hours."
"That the Conference expresses the hope that as soon as may be prac-
ticable the astronomical and nautical days will be arranged everywhere to
begin at mean midnight."
Although the other resolutions were adopted, the convention of west longitude
as positive was used in astronomy until 1984 when the general use of East posi-
tive was introduced. The Astronomical Almanac switched convention at that time,
exactly one hundred years after the International Meridian conference.
At the invitation of the Bureau des Longitudes, the directors of the national
ephemerides and other astronomers, met in Paris in May 1896 for the Conference
Internationale des Etoiles Fondamentales. In addition to adopting resolutions con-
cerning the fundamental catalog, and the calculation and publication of apparent
places of stars, the Conference adopted the following fundamental constants:
Nutation 9'.'21
Aberration 20'.'47
Solar parallax 8'.'80
It also agreed to adopt Newcomb's definitive values (which were not then in
final form) of lunisolar and planetary precession.
Active cooperation between the offices of the national ephemerides dates from
the Congress International des Ephemerides Astronomiques held at the Paris Ob-
servatory in October 1911. This conference was called by B. Bailaud, Director of
the Observatory and President of the Comite International Permanent de la Carte
13 / HISTORICAL INFORMATION g-jy
Photographique du Ciel on the initiative of the Bureau des Longitudes. Its pur-
pose was "d'etablir une entente permettant d'augmenter, sans nouveaux frais la
masse de donnees numeriques fournies annuellement aux observateurs et aux calcu-
lateurs." Although the Conference was primarily concerned with obtaining a greatly
increased list of apparent places of stars, it extended its attention to all the ephem-
erides of bodies in the solar system. Its comprehensive recommendations covered
the distribution of calculations between the five principal ephemeris offices (France
Germany, Great Britain, Spain, and the United States), specified standards of cal-
culation and presentation, arranged for publication of additional data, and fixed
the values of two further constants to be used in the ephemerides: the flattening
of the Earth (1/297) and the semidiameter of the Sun at unit distance for eclipse
calculations (15'59''63). Most of these recommendations are still in force.
Official approval was in some cases necessary for the adoption of these rec-
ommendations, as illustrated by the following extract from the Act of Congress of
August 22, 1912 (37 Stat. L., 328, 342):
"The Secretary of the Navy is hereby authorized to arrange for the exchange
of data with such foreign almanac offices as he may from time to time
deem desirable, with a view to reducing the amount of duplication of work
in preparing the different national nautical and astronomical almanacs and
increasing the total data which may be of use to navigators and astronomers
available for publication in the American Ephemeris and Nautical Almanac:
Provided..."
Here follows a number of provisions. The most important provision astronomi-
cally was the repeal of the proviso in the appropriation Act of September 28, 1850
(9 Stat. L., 513, 515) that "hereafter the meridian of the observatory at Washington
shall be adopted and used as the American meridian for all astronomical purposes,
and that the meridian of Greenwich shall be adopted for all nautical purposes."
Such exchange agreements have been carried out in spite of international diffi-
culties.
13.22 International Astronomical Union
In 1911, agreements had been directed almost entirely to the reduction of the total
amount of work by the avoidance of duplicate calculation. In 1938, Commission
4 recommended that this principle should be extended to the avoidance of dupli-
cate publication by the collection in a single volume of the apparent places of stars
then printed in each of the principal ephemerides. This recommendation, coupled
with the adoption of the Dritler Fundamentalkatalog des Berliner Astronomischen
Jahrbuchs (FK3), was implemented for 1941 by the publication of the international
EXPLANATORY SUPPLEMENT
volume Apparent Places of Fundamental Stars (under the auspices of the Interna-
tional Astronomical Union). This work gave astronomers access to the apparent
places of stars in one volume, and the individual ephemeris offices were saved the
work of the compilation and proofreading, as well as the cost of typesetting.
Continuing the precedents of the 1896 and 1911 conferences, the Director of
the Paris Observatory (Professor A. Danjon) convened a further conference that
was held in Paris in March 1950 to discuss the fundamental constants of astron-
omy. The leading recommendation was "that no change be made in the conven-
tionally adopted value of any constant." But the recommendations with the most
far-reaching consequences were those which defined ephemeris time and brought
the lunar ephemeris into accordance with the solar ephemeris in terms of ephemeris
time. These recommendations were addressed to the International Astronomical
Union and formally adopted by Commission 4 and the General Assembly of the
Union in Rome in September 1952.
Commission 4 had, at various times, made arrangements for the redistribu-
tion of calculations between the ephemeris offices. For example, the Institute for
Theoretical Astronomy in Leningrad contributed apparent places of stars to the
international volume for the years 1951-1959. With the availability of fast, au-
tomatic calculating machines, it was now both practicable and efficient for large
blocks of work, such as the calculation of apparent places of stars, to be done in
one office. At the 1955 General Assembly of the Union in Dublin, a general redis-
tribution of calculations along these lines was agreed upon by the directors of the
national ephemerides and confirmed by Commission 4. The reports of Commission
4 in Transactions of the International Astronomical Union documented the full
details of these agreements, the changes in the bases of the ephemerides, and the
discussions leading to the introduction of Apparent Places of Fundamental Stars.
In the Draft Report for the Twelfth General Assembly, held in Hamburg in
1964, D. H, Sadler indicated that the functions of Commission 4 on Ephemerides
were twofold: "firstly, to ensure that the published ephemerides fully meet the re-
quirements of astronomers and other users; and secondly, to coordinate the work of
the offices of the national ephemerides to ensure consistency, economy of effort, and
efficiency." Cooperation within the IAU on ephemerides has been directed toward
these functions. The primary emphasis of the first function has been the establish-
ment of a consistent and universal set of constants and bases for the ephemerides.
A brief review of the efforts in these two areas may be helpful.
At a meeting on fundamental constants for astronomy held in Paris during
March 1950, the definition of ephemeris time was recommended, and the lunar
ephemeris was brought into accordance with the solar ephemeris with respect to
ephemeris time. These recommendations were adopted in 1952 by Commission 4 of
the IAU at the Eighth General Assembly in Rome.
13 / HISTORICAL INFORMATION g
The IAU Symposium No. 21 (Paris, May 1963) concluded that a change in
the conventional IAU system of constants could no longer be avoided. The incon-
sistencies and inadequacies of the system of that time, the better values of some
constants from recent determinations, and deficiencies revealed by discussions of
high accuracy observations indicated the need for new constants.
At the Twelfth General Assembly (Hamburg, 1964) a working group reviewed
the report of the IAU Symposium. A list of constants proposed by the Working
Group on the System of Astronomical Constants was adopted and recommended
for use at the earliest practicable date in the national and international astronomical
ephemerides. These constants were introduced in the ephemerides for 1968. It was
also noted that the constants of precession and planetary masses had not been
changed and that consideration should be given to their future improvement.
In August 1970, IAU Colloquium No. 9 on the IAU System of Astronom-
ical Constants was held in Heidelberg, and recommended the establishment of
three working groups — on planetary ephemerides, on precession, and on units and
timescales. The recommendations were adopted and the working groups were es-
tablished at the 1970 IAU General Assembly in Brighton, England. A working
meeting on constants and ephemerides was held in October 1974, in Washington,
DC to draft a proposed report of the working groups. The chairmen of the working
groups met in September 1975, and June 1976, in Herstmonceaux and Washing-
ton, respectively. The Report and Recommendations, known as the Joint Report of
the Working Groups of I A U Commission 4 on Precession, Planetary Ephemerides,
Units and Time Scales were adopted by the IAU in August 1976 in Grenoble.
In 1976, a working group on cartographic coordinates and rotational elements
of planets and satellites was established by IAU Commissions 4 and 16. In 1977, a
working group on nutation was established by Commission 4. These groups were ex-
pected to provide recommendations for consideration at the 1979 General Assembly
in Montreal.
The second primary function of Commission 4 on ephemerides — the coordina-
tion of the efforts of the various offices — was being fulfilled with the move from the
distribution of calculations among the separate offices, the unification of printing,
and the exchange of reproduction proofs for printing by the various countries. This
change was made possible by the advent of high-speed computers.
At the Fourth General Assembly of the IAU held in Cambridge, MA in 1932,
Dr. L. J. Comrie, Director of the British Nautical Almanac Office, suggested that
duplicate printing in the national volumes of ephemerides be discussed, particu-
larly with respect to the apparent places of stars. At that time, Professor Herrero
suggested that the ideal would be an international almanac. At the Fifth General
Assembly (Paris, 1935) an agreement was reached for a single publication of the
Q2Q EXPLANATORY SUPPLEMENT
Apparent Places of Stars, initially to be printed in Great Britain and now printed
in Germany.
At the Ninth General Assembly (Dublin, 1955), an international fundamental
astronomical ephemeris (IFAE) was discussed. The IFAE was to be a single pub-
lication, under the auspices of the IAU, containing the fundamental astronomical
ephemerides to the fullest accuracy. The national ephemerides could then be much
smaller and cater more directly to the practical astronomer. Although many were in
favor of this proposal, several factors prevented its adoption: the practical difficulty
that the sales were not likely to cover the cost of printing, because of the antic-
ipation of required, free distribution; the difficulty for astronomers of purchasing
books published in other countries; the required cooperation of almost all national
ephemerides; and the loss of flexibility. However, various countries using reproduc-
tion proofs of material prepared by a single source was an attractive alternative.
At the same time, it was announced that an agreement had been reached for the
unification of The American Ephemeris and Nautical Almanac and The Nautical
Almanac and Astronomical Ephemeris beginning with the year 1960. It was hoped
that other ephemeris offices would save considerable composition and proofreading
costs by using photolithography for reproduction.
At the Tenth General Assembly (Moscow, 1958) it was reported that the
Astronomisch-Geodatisches Jahrbuch, introduced for the year 1949, would cease
publication after the 1959 edition and that the Berliner Astronomisches Jahrbuch,
introduced for the year 1776, would cease publication after the edition for 1959.
As a result of these savings of composition and printing costs, the Astronomisches
Rechen-Institute was able to take over composition and publication of the Appar-
ent Places of Fundamental Stars. Although it was sad to see a publication of long
standing cease, the change indicated an increased international cooperation and
permitted a beneficial transfer of functions.
G. M. Clemence and D. H. Sadler reported at the Eleventh General Assembly
(Berkeley, 1961) that an important step towards unification of the national ephem-
erides had taken place in the unification of the American and British volumes. They
stated further that:
"Many dearly-held, but essentially unimportant, standards and prejudices
have had to be sacrifices on both sides; it is surprising how quickly these lose
their former importance in the satisfaction of a comprehensive agreement.
That same cooperation, good will and confidence exists between all the
national ephemeris offices, and, although differences of language will intro-
duce some further difficulties, there is no obstacle to complete unification
that will not be overcome in course of time."
Between 1962 and 1964 the Japanese and Soviet ephemerides began using the
advanced proofs from The Astronomical Ephemeris.
13 / HISTORICAL INFORMATION ggi
In October 1974, plans were begun to revise the organization, content, and basis
for The American Ephemeris and Nautical Almanac/Astronomical Ephemeris. It
was decided that a single unified printing in English would be made in the United
States, and that it would be available from both Her Majesty's Stationery Office
in England and the Superintendent of Documents in the United States. Since the
United States legal code required the title The American Ephemeris and Nautical
Almanac, a bill was introduced into Congress to modify the legal code to permit a
change in the name of the publication. Also, it was decided that the organization
and content would be changed for the 1981 edition.
1 3.23 Other International Organizations
Improvements in accuracies and in knowledge concerning the motions and rotations
of the bodies in the solar system, particularly the rotation of the Earth and our
use of accurate time, has created a growing need for closer cooperation between
the different international organizations. Thus, many international organizations
become involved in various aspects of international standardization. Some of these
different organizations and their functions are described in the following paragraph.
The International Association of Geodesy (IAG) is interested in the rotational
elements and physical characteristics of the planets. The Committee on Space Re-
search (COSPAR) is interested in the cartographic coordinate systems and the
gravity fields of the planets and satellites. The Consultative Committee on Inter-
national Radio (CCIR) is responsible for standardization of time and frequency
as used in radio transmissions; its Study Group Seven reaches agreements in this
area. The Committee Internationale des Poides et Mesures (CIPM) coordinates
the standards of the units of measure. The Consultative Committee on the Def-
inition of the Second (CCDS) is responsible for the definition of the second and
the principles for measuring the second. The Committee on Data for Science and
Technology (CODATA) works on an interdisciplinary basis to improve the quality,
reliability, processing, management, and accessibility of data of importance to sci-
ence and technology. The Bureau Internationale des Poides et Measures (BIPM) in
Paris, France, is responsible for maintaining International Atomic Time (TAI) and
providing a single atomic timescale, worldwide. The International Earth Rotation
Service (IERS) determines the rotation of the Earth, and thus the UTC timescale.
It also determines when leap seconds should be inserted into the UTC timescale.
13.3 HISTORICAL LIST OF AUTHORITIES
13.31 Introduction
It is important to know the basis of the ephemerides to properly interpret the results
of past discussions of observations. For this reason, and for purposes of historical
§22 EXPLANATORY SUPPLEMENT
interest and record, the following list of authorities is given in the original form. In
addition, a few uncertainties and ambiguities, which could only have been resolved
by excessive research or recalculation, have been allowed to remain.
The material is arranged in the following subgroups: up to 1900 The Nautical
Almanac and The American Ephemeris are treated separately, and after 1900 they
are combined. Each subsection is divided according to body or subject (e.g., Sun,
Moon, precession, nutation, satellite constants). Within each division the author-
ities, arranged chronologically, are preceded by a short narrative of the quantities
tabulated. In these discussions "and" is used to indicate that the quantity is tabu-
lated to n decimal places, and the term "precision" is used to indicate merely the
unit of the end figure. Usually, only names and dates are given for the unit of the
end figure. Also, only names and dates are given for the authorities; full references
are given in Section 13.4. In some cases, for example for the adopted semidiameters
of the planets, detailed references to the original publications have been omitted.
Some of the tabulated ephemerides are based on theories, derivations, and
constants given in appendices and supplements to The Nautical Almanac and to
The American Ephemeris. Lists of these appendices and supplements are given in
Section 14.6.
1 3.32 The Nautical Almanac (1 767-1 900)
13.321 Sun All ephemerides from 1767 to 1833 were given with the argument ap-
parent time. Quantities tabulated for the Sun at intervals of one day were longitude
and declination (each to 1"), and right ascension and equation of time (each to l5).
Semidiameters in arc (to 0"1), in time (to 0!1), and log distance (to 6D) were given
at intervals of 6 days. From 1768, the equation of time (to 0?1) was given, and from
1772, the right ascension (to 0!1) was also given. For the years 1815-1822, the log
distance (to 5D) was given. In 1833, the semidiameters (to 0"01, to 0!01) and log
distance (to 7D) were given at intervals of one day.
The 1834 almanac was largely remodeled in accordance with the Royal Astro-
nomical Society's report printed in that almanac, and thereafter the argument of
most ephemerides was mean time. Most quantities tabulated in time were given
to 0?01, those in arc to 0"1 (except the Sun's latitude, given to Ol'Ol), and the log
radius vector to 7D. In many cases, differences or variations were given. From 1848,
equational rectangular coordinates (to 7D) were included at intervals of one day,
latitude terms being included for the first time in 1866; the values for 1845-1847
were given in the 1848 volume. No other substantial changes were made before
1900.
1767-1796: Mayer's "last manuscript tables" assumed an annual precession of
50"3.
13 / HISTORICAL INFORMATION
1797-1804: Mayer's tables with the mean motion corrected to the revised preces-
sion of 50"2.
1805-1812: Delambre's tables as given in Lalande (1792), but with certain (un-
specified) coefficients determined by Maskelyne.
1813-1821: Improved tables by Delambre (1806).
1822-1832: The tables in Vince (1808, Volume III), "with the omission only of
some equations which do not materially effect the results." The tables
are stated by Vince (Volume III, page 2) to have been "constructed
by M. de Lambre, from observations of Dr. Maskelyne, and the theory
of M. Laplace. See Les Memories de Berlin, for 1784, 1785." (In 1832
the position of the Sun for the calculation of the transit of Mercury
was taken from Carlini's tables.)
1833: The longitude was taken from Delambre's tables, improved by Airy's
corrections based on Greenwich observations.
1834-1835: Carlini's tables (1810) with Bessel's corrections (1828) and nutation
as in the Astronomical Society's tables (Baily, 1825). The elements
used by Carlini are the same as those of Delambre (1806), but the
arrangement is better for the construction of an ephemeris.
1836-1863: Carlini (1832).
1864-1900: Leverrier (1858).
13.322 Moon The Moon's longitude, latitude, semidiameter, and horizontal par-
allax (each to 1") and its right ascension and declination (each to 1') were tabulated
at intervals of 12h (apparent time) for the years 1767-1833. Lunar distances (from
at least one star, and after 1770 from one or two stars as well as the Sun when the
Sun was conveniently placed) were given to a precision of 1" for every 3". From 1823,
the right ascension and declination were given to 1". In 1834, the argument became
mean time and, with occasional minor alterations, the tabulations were given to an
extra figure until the year 1900. The right ascension (to 0!01) and the declination
(to 0'.'1) were given at intervals of one hour.
1767 1776: Mayer last manuscript tables.
1777-1 788: Mayer's tables, improved by Mason under Maskelyne's direction, based
on Bradley's observations (the latter are printed in The Nautical Al-
manac for 1774).
1789 1796: Mayer's tables were further improved by Mason (1780). Eight new
equations were taken from Mayer's tables, the coefficients being de-
termined from Bradley's observations. The eighteenth equation in lon-
gitude was omitted.
1797-1804: The same set of tables, but adjusted (as for the Sun) for the corrected
value of precession.
524 EXPLANATORY SUPPLEMENT
1805-1807: Lalande (1792); the tables are the same as Mason (1780) except for
the substitution of Laplace's acceleration and secular motion.
1808: Lalande's tables, with the addition of two further inequalities found
by Laplace.
i809 1812: The epochs, Laplace's accelerations, and "a particular equation of his"
were taken by Maskelyne from Burg's tables, and the mean longitudes
were computed. The parallax was taken from Mayer.
1813-1817: Burg (1806) on Laplace's theory, the coefficients being determined
from Maskelyne's observations, and the epochs from those of Maske-
lyne and Bradley.
1818-1820: According to Pond's preface, the tables of Burckhardt were used. [But
a note (initialed T.Y.) at the end of the 1820 preface states that those
of Burg were used.]
1821-1833: Burckhardt (1812).
1834-1855: Burkhardt's tables, with nutation from Baily (1825).
1856: As in the previous years, but the parallax taken from Adams (1853b)
and the semidiameter taken as 0.2725 times the horizontal parallax.
1857-1861: The ratio of semidiameter to horizontal parallax was changed to
0.273114.
1862-1882: Hansen (1857).
1883 1895: Hansen, but with Newcomb's corrections (1878b) included in the right
ascension and declination.
1896: As in previous years, and with the substitution of Newcomb's Table
XXXIV for Hansen's.
1897-1900: Newcomb's corrections included in horizontal parallax and semidiam-
eter.
13.323 Major Planets Ephemerides of the five "classical" planets were given at
intervals of 6 days until 1832, except for Mercury, which were given at intervals of 3
days from 1778. Those of Uranus (at intervals of 10 days) were introduced in 1789
and again from 1791 onwards. The adopted precision was 1" for declination and both
heliocentric and geocentric longitudes and latitudes. When the right ascension was
added, in 1819, a precision of lm was used.
Heliocentric coordinates were omitted in 1833, while declination (to 1"), geo-
centric longitude and latitude (to 1"), right ascension (to 0!1), and log distance (to
5D), were all given at intervals of one day.
The intervals were changed in 1834 to one-day intervals for all planets. In 1861,
when Neptune was introduced, the intervals for Uranus and Neptune were changed
to 4 days. Right ascension (to 0!01), declination (to O'.'l), heliocentric longitude
and latitude (to Ol'l), and log distance and log radius vector (each to 7D) were
included. Geocentric longitude and latitude were omitted for the year 1834. A geo-
centric (equatorial) ephemeris of Neptune was published between 1850 and 1860,
13 / HISTORICAL INFORMATION g2
at intervals of 5 days, usually as an appendix to later Almanacs.
Transit ephemerides were introduced in 1839 for Mercury to Uranus, and i
1861 for Neptune.
Mercury, Venus, Mars, Jupiter, Saturn
1767-1779: Halley (1749).
1780-1804: Wargentin's tables, "annexed to M. De Lalande's Astronomy."
1805-1833: Lalande (1792). These are the tables calculated by Delambre on the
theory of Laplace. From 1822, the tables of Mars were taken from
"those of Lalande in the Connaissance des Terns [sic] for the 12th
year [1803-04]." The pi aces of Mercury for the transit of 1832 were
taken from Lindenau's tables.
Mercury
1834-1863: Lindenau (1813).
1864- 1900: Leverrier (1859).
Venus
1834-1864: Lindenau (1810). For the years 1837-1848 a correction of -2' 18" was
applied to the tabular longitude of the node.
1865- 1900: Leverrier (1861a).
Mars
1834-1865: Lindenau (1811).
1866- 1900: Leverrier (1861b).
Jupiter
1834-1877: Bouvard (1821).
1878 1900: Leverrier (1876a).
Saturn
1834-1879: Bouvard (1821). [For the years 1852-1879, Bouvard's Table 42 was
used in the corrected form given by Adams (1849) and in The. Nautical
Almanac 1851, xiv.]
1880-1900: Leverrier (1876b).
g26 EXPLANATORY SUPPLEMENT
Uranus ("The Georgian" in The Nautical Almanac 1789-1850)
1789, 1791-1833: Computed from the same tables used for the "classical" planets.
1834-1876
1877-1881
1882-1900
Bouvard (1821).
Newcomb (1873).
Leverrier (1877a).
Neptune
1850-1857: Computed from elements given in various issues of the Berliner Jahrbuch
and The Nautical Almanac.
1858-1870: Kowalski (1855). (This is a little uncertain for the years 1859-1860,
as the supplements to the almanacs containing these ephemerides do
not quote the authority.)
1871-1881: Newcomb (1873).
1882-1900: Leverrier (1877b).
13.324 Minor Planets Ephemerides of minor planets were given for the first time
in the almanac for 1834. That issue contained ephemerides based on elements by
Encke, at intervals of 4 days throughout the year for Ceres, Palles, Juno, and Vesta.
The right ascension was tabulated to 0T1, declination and heliocentric longitude and
latitude to 1', and log distance and log radius vector to 4D. For one month on each
side of opposition, at intervals of one day, the right ascension was given to 0!01,
declination to O'.'l, and log distance and log radius vector to 5D.
Similar ephemerides for the years to 1849 were based on the same elements,
with variations calculated by the method given by Airy (1835).
Between 1850 and 1866, the number of planets for which ephemerides at wider
intervals were published, was increased to as many as 36 in some years. The elements
used were from a number of different authors. (The almanac for 1856 contains a
translation (by Airy) of papers by Encke (1852a and 1852b) on the computation of
special perturbations.)
From 1867, ephemerides for only live planets were published, and from 1876,
only the first four planets were included, on the grounds that more accurate ephem-
erides were to be published in the Berliner Jahrbuch.
Ceres
1867-1881: Schubert (1854).
1882-1900: Godward (1878).
Pallas
1867-1900: Farley (1856a).
13 / HISTORICAL INFORMATION
Juno
627
1867-1893: Hind (1855).
1894-1900: Hind (1855), with corrections by Downing (1890).
Vesta
1867-1900: Farley (1856b).
Astraea
1867-1875: Farley (1856).
13.325 Satellites No predictions for satellites, except for the Galilean satellites of
Jupiter, were published until 1899. Diagrams of the apparent orbits at the time of
opposition and elongations (to 0?l) for a limited period around opposition, were
given for the satellites of Mars, Saturn, Uranus, and Neptune. The authorities for
the latter are the same as that for 1901, and are listed in Section 13.345.
Diagrams of the configurations and predictions of eclipses for Jupiter's satellites
have been published in every issue of The Nautical Almanac.
1767-1804
1805-1823
1824-1833
1834-1839
1840-1900
Eclipses to Is; based on Wargentin (1746).
Based on Lalande (1792) (quoting Delambre).
Based on Delambre (1817).
Eclipses to 0?1. Other phenomena to lm; based on Delambre (1817).
As in 1834 1839. From 1877, eclipses to Is. From 1896, times of con-
junction to OF I, based on Damoiseau (1836), and extensions by Adams
and others.
13.326 Auxiliary Quantities Auxiliary quantities include sidereal time, mean obliq-
uity of the ecliptic, apparent obliquity of the ecliptic, precession, nutation, con-
stants, and semidiameters at unit distance.
13.3261 Sidereal time The sidereal time was not tabulated explicitly for the first
sixty or seventy years, but from 1833, values were given at intervals of one day. The
sidereal time at mean noon is stated to have been calculated from the following
expressions:
1833: Sun's mean longitude + 6'.'0 - 1 6'.'5 sin Q - 0'.'9 1 7 sin 2
1834 1900: Sun"s mean longitude + nutation, where the Sun's mean longitude
at Paris mean noon of January 0 of the year 1800 + t is given by
Bessel (1830a, p. xxiv) as 279° 54' 0 1 '.'36+27 "605 844 r+0"000 1 22 1 805 11 —
g28 EXPLANATORY SUPPLEMENT
I4'47'.'083/. where / is (for the 19th century) the number of years from
the preceding leap year.
13.3262 Mean obliquity of the ecliptic The following values were used for the years
1767-1900 (/ being measured in years):
1767-1807: 23°28' l6"-x(t- 1756). Mayer (1770, pp. 105 and v). Maskelyne stated
in several almanacs that x was about half a second, but Mayer's table
indicates 0"46. The values seem to have been adjusted occasionally
by Maskelyne.
1808-1833: Corrected year by year, from Greenwich observations to a current
date.
1834-1863: 23°27'54"8 - 0'.'457(r - 1800. 0). Bessel (1830a, p. xxvii).
1864-1900: 23°27'31"83 - 0'.'476(/ - 1850. 0). Leverrier (1858, p. 203).
The authorities for the values of the obliquity adopted for the conversion of the
Moon's longitude and latitude to right ascension and declination were (see A.M.W.
Downing, M.N.R.A.S., 69, 618, 1909):
1862 1874: Hansen (1857, p. 45: see Hansen and Olufsen, 1853, p. 5).
1875 1900: Leverrier (1858, p. 203).
Reducing all these to a common date, for comparison with Peters (1842) and
Newcomb (1895a), we have:
Mayer 23°27'29" - 0"46(r - 1850. 0).
Bessel 23°27,31'.,95 - 0"457(f - 1850.0).
Leverrier 23°27'31"83 - 0'.'476(r - 1850.0).
Hansen and Olufsen 23°27'3r.'42 - 0'.'46784(f - 1 850. 0).
Peters 23°27'30"99 - 0'.'4645(f - 1850.0).
Newcomb 23°27'3r.'68 - 0'.'468(r - 1850.0).
13.3263 Apparent obliquity of the ecliptic Values of the apparent obliquity (toOl'l)
at intervals of three months were published from the inception of the Almanac; in
1817 and 1818, and again from 1834, the interval was changed to 10 days, and in
1834, the precision was changed to O'.'Ol. From 1876 to 1895, the short-period terms
of nutation were included in the apparent values, which were tabulated at intervals
of one day.
13.3264 Precession Mayer's value (1770, p. 52) of 50"3 was used in the almanac
from 1767 to 1796, but was corrected to 50'.'2 from 1797 to 1833, with a corre-
sponding adjustment of Mayer's values of the mean motions of the Sun and Moon.
13 / HISTORICAL (NFORMATION g29
Between 1834 and 1853 there is no specific statement of the values used, but from
1854 to 1895 the annual (and daily) increments were given (to O'.'OOOl), and the
precession from the beginning of the year was tabulated (to Ol'Ol) at intervals of
10 days. No authority is quoted for these figures. From 1896 to 1900, Peters' value
(1842, p. 71) is stated to have been used as the authority.
The following comparison shows the various values of the annual precession
that were used (T being measured in centuries from 1850.0):
1854-1856 (deduced): 50"2357 + 0"025 T.
1857-1895 (deduced): 50'.'2524 + 0?227 T.
1896 1900 (Peters): 50"2524 + 0^227 T.
1901-1959 (Newcomb): 50:'2453 + 0"0222 T.
13.3265 Nutation Values for nutation have appeared since 1767. However, the terms
included have changed over the years.
1767 1833: The "Equation of the Equinoctial Points" (nutation in longitude) was
tabulated (to O'.'l) for every three months, but without any indication
of the authority or of the terms included. In the years 1817 and 1818,
the same equation in sidereal time (or nutation in right ascension)
was given (to 0!01) at intervals often days.
1834-1856: The tabulated values in longitude (to Ol'Ol ) and in right ascension (to
0?01 ) at intervals of 10 days, were based on Baily's (1825) values and
included the four terms numbered I, 2, 3, 14 in Table 13.348.1 (page
654).
1857 1880: The same terms were tabulated, and the precision and the interval of
tabulation were unchanged, but the coefficients were based on Peters
(1842).
1881-1892: Nutation in obliquity was also included, to a precision of 0"01.
1893-1895: Two additional terms (5 and 7) were included in the tabulations.
1896: Long-period and short-period terms in both longitude and obliquity
were included in the tabulations, the interval of which was one day.
Term no. 15 was included.
1897-1900: Nine additional terms (6, 8-13, 17, 18) were included.
13.3266 Constants The following values of the principal constants have been used:
Solar parallax
1834-1869
1870-1881
1882-1900
8:'5776 Encke (1824, p. 108).
8'.'95 Leverrier (1858, p. 114).
8'.'848 Newcomb (1867, p. 29).
630
Constant of Aberration
EXPLANATORY SUPPLEMENT
1834-1849: 20"36
1850-1856: 20'.'42
1857-1900: 20"4451
Baily {1825, p. x).
Baily (1845, p. 21).
Struve (1844, p. 275),
Constant of nutation
1834-1856: 9?25 Baily (1823, p. xiv).
1857-1900: 9'.'223 1 + 0!'0009 t, where T is in centuries from 1800.0 (Peters, 1842,
p. 75).
13.3267 Semidiameters at unit distance Semidiameters are given for the Sun, Mer-
cury, Venus, Mars, Jupiter (equatorial and polar), Saturn (equatorial and polar),
Uranus, and Neptune.
Sun
1767-
-1807
962'.'8
Mayer (1770, p. 56).
1808-
-1833
961"37
Mayer (1770, p. 56).
1834-
-1852
960'.'9
Bessel (1830a, p. L).
1853-
-1895
961'.'82
Airy (1855, p. xxviii).
1896-
-1900
961 "18
Auwers (from observations at Greenwich, 1851-1883).
1896-
-1900
959"63
Auwers (1891, p. 367) (This value was used for eclipse
calculations only.)
Mercury
1834-1863: 3'.'23 Lindenau.
1864-1900: 3'.'34 Leverrier.
Venus
1834-1864: 8"25 Delambre.
1865-1895: 8'.'305 Leverrier.
1896-1900: 8'.'40 Auwers.
Mars
1834:
1835-1865:
1866-1895:
1896-1900:
4'.'57.
4'.'435
5 '.'5 5 [sic]
4'.'68
Littrow.
Leverrier.
Hartwig.
13 / HISTORICAL INFORMATION
Jupiter (equatorial)
631
1837-1881: 991704 Struve.
1882-1895: 98"19 Leverrier.
1896-1900: 97'.'36 Schur.
Jupiter (polar)
1834: 931'37.
1835-1856: 93"4
1857-1881: 92"426
1882-1895: 92"200
1896-1900: 91'.'10
Delambre.
Equatorial SD X 0.927.
Equatorial SD x 0. 939.
Schur.
Saturn (equatorial)
1834: 88".72.
1837-1881: 81"106 Bessel.
1882-1895: 83:'31 Leverrier.
1896-1900: 84'.'75 Meyer.
Saturn (polar)
1835-1856: 75'.'25
1857-1881: 75['19
1882-1895: 74'.'56
1896-1900: 6'.'88
Bessel.
Equatorial SD X 0. 927.
Equatorial SD x 0. 895.
Meyer.
Uranus
1834: 37['20.
1835-1881: 37'.'25 Delambre.
1882-1895: 34'.'28 Leverrier.
1896-1900: 34!'28 Hind.
Neptune
1899-1900: 34!'56 Barnard; however, no values were tabulated.
1 3.33 The American Ephemeris (1 855-1 900)
The twelve monthly sections that formed the principal content of the first part of
The American Ephemeris contained the Greenwich ephemerides of the Sun and
632 EXPLANATORY SUPPLEMENT
Moon, which remained virtually unchanged during the period 1855 to 1900. The
Greenwich ephemerides of Venus, Mars, Jupiter, and Saturn; similar ephemerides of
Mercury, Uranus, and Neptune; and heliocentric ephemerides for all seven planets
were added in 1882. An ephemeris of the rectangular coordinates of the Sun was
also included, although for the period 1875-1881, it was relegated to the second
part of The American Ephemeris.
The second part of the book contained the ephemeris for the meridian of Wash-
ington, and included further ephemerides of the Sun, Moon, and planets, partly for
Washington noon and midnight and partly for meridian transit at Washington.
These ephemerides, which were revised somewhat from time to time, included a
tabulation of the obliquity, precession, and nutation until 1882, when it was trans-
ferred to the first part of the volume.
13.331 Sun In the Greenwich ephemerides of the Sun, the apparent right ascen-
sion (to 0?01), the declination (to 0"1), and the equation of time (to 0!01) were
tabulated at intervals of one day for apparent noon and (except in 1855) for mean
noon. The semidiameter (to O'.'Ol) and the sidereal time of semidiameter passing
the meridian (to 0*01 ) were given for apparent noon. Except in 1855, the longi-
tude (to O'.'l) referred both to the true equinox of date and to the mean equinox
of the beginning of the year. The latitude (to O'.'Ol), and the log radius vector (to
7D) were given for mean noon. In the Washington ephemerides, the right ascension
and declination were given for mean and apparent noon. The equation of time,
the semidiameter, and the sidereal time of semidiameter passing the meridian were
given for apparent noon. From 1855 to 1881, the longitude, latitude, and log radius
vector were given for Washington mean noon and midnight. The tabular precisions
were the same as that in the Greenwich ephemerides.
During the period 1855-1881, the equatorial rectangular coordinates of the
Sun (to 7D) tabulated for Greenwich mean noon, referred to the true equinox and
equator of date. For Washington, mean noon and midnight, referred both to the
true equinox of date and the mean equinox of the beginning of the year. In 1882,
these ephemerides were replaced by a tabulation (to 7D) for Greenwich mean noon
and midnight, which referred both to the true equinox of date and mean equinox
of the beginning of the year.
The horizontal parallax (to O'.'Ol) and aberration (to O'.'Ol) of the Sun were
tabulated at intervals of 10 days, for 0" Washington sidereal time from 1855 to
1864, for Washington mean noon from 1865 to 1881, and for Greenwich mean noon
thereafter.
1855-1857: Carlini (1810), with Bessel's revisions (1828).
1858-1874: Hansen and Olufsen (1853).
1875-1899: Hansen and Olufsen (1853), with aberration according to Struve (1844).
1900: Newcomb (1895a).
13 / HISTORICAL INFORMATION 533
13.332 Moon In the Greenwich ephemerides of the Moon, the right ascension (to
0!01) and declination (to O'.'l) were tabulated for every hour; the semidiameter and
horizontal parallax (to O'.'l) were given for noon and midnight; and, from 1860 to
1900, the longitude and latitude (to 0!'l) were given for noon and midnight.
In the Washington ephemerides, the right ascension (to 0!01) and declination
(to 0'.'1) were tabulated for upper and lower culmination from 1855 to 1864, and for
upper culmination from 1882 to 1900. The sidereal time of semidiameter passing
the meridian (to 0!01) was given for both culminations during the period 1855-
1864, but only for upper culmination after 1864. The semidiameter and horizontal
parallax (to 0"1) were tabulated for Washington mean noon and midnight from
1855 to 1881, and for upper culmination from 1882 to 1900.
The times of the phases, apogee, and perigee were given both in Greenwich
mean time and in Washington mean time. The mean longitude and the longitude
of the ascending node were given at intervals of 10 days.
1855-1856: Pierce (1853). These tables are based on Airy (1848), with corrections
by Airy (1849) and Longstrength (1853). The tables used by Airy
(1848) were derived from Damoiseau (1824), and were substantially a
development of Plana's theory (1832), modified to include two Venus
inequalities discovered by Hansen (1847).
1857-1882: Pierce (1853), derives parallax from tables based on formulas of Adams
(1853a) and Walker (1848),
1883-1900: Hansen (1857) contains corrections by Newcomb (1878b). [In the In-
troduction to The American Ephemeris for 1912 and following years,
attention was called to the fact that these corrections were not pre-
cisely in accordance with the statement given in the volumes for 1883-
1911, and the formula actually used was given.]
13.333 Major Planets After 1881, in the Greenwich ephemerides of the planets,
the apparent right ascension (to 0?01) and declination (to O'.'l) were tabulated for
mean noon at intervals of one day for Mercury, Venus, Mars, Jupiter, and Saturn,
and at intervals of 4 days for Uranus and Neptune. The time of meridian passage
(to 01" 1) was also given. The semidiameter and horizontal parallax (in general, to
0"1, and to 0"01 for Uranus and Neptune in 1900, to 0!'01 for Mercury, Venus, and
Mars) were tabulated at various different intervals for different planets.
In the Washington ephemerides, values were tabulated for the apparent right
ascension (to 0*01) and declination (to Oil), for the inferior planets at Washington
mean noon and meridian transit, and for the superior planets at Washington sidereal
noon and meridian transit between 1855 and 1869. After 1870, mean noon and
meridian transit time were used for all the planets, but after 1882 the ephemerides
for noon were omitted. The semidiameter and horizontal parallax (to O'.'Ol) were
given for 0h Washington sidereal time from 1855 to 1864, for Washington mean
534 EXPLANATORY SUPPLEMENT
noon from 1865 to 1881, and (to O'.'l) for Washington meridian transit beginning
with 1882. The sidereal time of semidiameter (to 0101) passing the meridian was
tabulated throughout.
Included in the second part of the volume from 1855 to 1881 were also helio-
centric ephemerides of the planets. The quantities tabulated were the rectangular
coordinates (to 4D) for Mercury, Venus, Mars, and Neptune; to 5D for Jupiter, Sat-
urn, and Uranus, and the orbital longitude (to O'.l for the inner planets; to 1° for
the outer planets). The attraction on the Sun was added in 1961. During the period
1855-1860, the rectangular coordinates were referred to the equinox and equator
of date; beginning with 1861, they were referred to the ecliptic and mean equinox
of a selected epoch, and the coordinates of the Earth were included. A table of the
adopted masses and the orbital inclinations and nodes was also given.
In 1882, the heliocentric ephemerides were replaced by different ones and fol-
lowed immediately after the geocentric ephemerides in Part I. These heliocentric
ephemerides contained the longitude and latitude (to O'.'l) referred to the ecliptic
and mean equinox of date. Also included were the reduction to orbit, log radius
vector (to 7D); in 1900, to 8D for Venus and Mars, and log geocentric distance
(to 7D), for Greenwich mean noon at intervals of 8 days for Uranus and Neptune.
2 days for Mercury (one day in 1900), and 4 days for the other planets (2 days
for Venus and Mars in 1900). Log geocentric distance was also given for the dates
between the tabular dates.
During the years preceding the completion of the planetary tables of Newcomb
and Hill, the ephemerides were calculated with tables that, for the most part, were
constructed by applying corrections to the early tables of Lindenau and Bouvard
that were based upon Laplace's theories.
Mercury
1855-1899: Winlock (1864), based on the theory of Leverrier (1845).
1900: Newcomb (1895b).
Venus
1855-1875: Manuscript tables prepared from Lindenau (1810) by applying cor-
rections based on investigations by Airy (1832), Breen (1848), and
Leverrier (1841).
1876-1899: Hill (1872). (An ephemeris for 1874-1875 calculated from these tables
is given in the Appendix to the 1876 volume.)
1900: Newcomb (1895c).
13 / HISTORICAL INFORMATION
635
Mars
1855-1899: Manuscript tables based on Lindenau (1811), with corrections from
Breen (1851) and Leverrier (1841), and various other corrections from
time to time.
1900: Manuscript tables, based on the elements derived by Newcomb (1895d).
Jupiter
1855-1897: Manuscript tables, prepared from Bouvard (1821), with corrections
to make them agree with observation.
1898-1900: Hill (1895a). (In the 1898 volume, it is incorrectly stated that Bou-
vard's tables were used for that year.)
Saturn
1855-1882: Manuscript tables prepared from Bouvard (1821), with various cor-
rections from time to time.
1883-1899: Manuscript tables prepared from a provisional theory by Hill (1890).
1900: Hill (1895b).
Uranus
1855-1875: Bouvard (1821), with revisions by Leverrier (1846) and Peirce (1848a),
and beginning with 1859, further corrections by Runkle (1855).
1876: Manuscript tables constructed by Newcomb.
1877-1900: Newcomb (1873). (An ephemeris for 1873-1876 calculated with these
tables is given in the 1877 volume.)
Neptune
1855-1869: Tables based on Peirce's theory (1848b) and Walker's elements (1848).
(Ephemerides for 1853 and 1854 were given in the volumes for 1855
and 1856, respectively.)
1870-1900: Newcomb (1865). (An ephemeris for 1866-1869 calculated from these
tables is given in the Appendix to the 1869 volume.)
13.334 Minor Planets Among the early tables constructed and printed for The
American Ephemeris were tables of the minor planets ( 15) Eunomia. (40) Harmonia,
(18) Melpomene, and (11) Parthenope. Ephemerides of these minor planets were
not included in The American Ephemeris, but an "Asteroid Supplement" in the
volume for 1861 contained opposition ephemerides for 33 minor planets for 1859,
and the orbital elements of (l)-(55).
636 EXPLANATORY SUPPLEMENT
13.335 Satellites In the volumes for 1855-1881, ephemerides were given only for
the four great satellit es of Jupiter, and the apparent, elements of the rings of Saturn.
The ephemerides of the satellites of Jupiter gave the superior geocentric conjunc-
tions, the phenomena, the coordinates in the mean apparent eclipses, and diagrams
of the phases of the eclipses.
In 1882, diagrams of the configurations of the four satellites of Jupiter were
added, and the former ephemerides of the coordinates in the apparent orbits were
omitted. A diagram of the apparent orbits was also added. Ephemerides of the
elongations of the satellites of Mars, Saturn, Uranus, and Neptune, and diagrams
of the apparent orbits, were introduced.
The elongations of the fifth satellite of Jupiter were added in 1898, but no
statement of authority was given from 1898 to 1900. No authority was given for the
satellites of Mars from 1882 to 1900. The authorities for the other satellites were:
Jupiter
L8SS-1881: Damoiseau (1836). Extended to 1880 by Kendall (1877).
1882-1900: For elongations and eclipses, Todd's (1876) continuation of Damoi-
seau; for occupations, transits, etc., Woolhouse (1833), with Table II
for each satellite adapted to Damoiseau.
Saturn
1855-1900: For rings, except the dusky ring, Bessel (1875a and 1875b).
1882-1900: For satellites, manuscript tables prepared by Newcomb.
Uranus and Neptune
1882-1900: Newcomb (1875).
13.336 Auxiliary Quantities
13.3361 Sidereal time The sidereal time (to 0!01 ) was tabulated for every Wash-
ington mean noon, and (except in 1855) for every Greenwich mean noon. In the
Greenwich ephemeris, except in 1855, the mean of 0h sidereal time (to 0?01) was
also given for every day.
13.3362 Obliquity of the ecliptic The adopted expressions and authorities for the
mean obliquity were:
1855-1881: 23°27'54,.,22 - 0:,4645 f-O'.'OOOOOUf2, where t is reckoned in years from
1800 (Peters, 1842).
13 / HISTORICAL INFORMATION 537
1882-1899: 23°27'3 1 ,.'42-0,.'46784 r, where t is reckoned in years from 1850 (Hansen
and Olufsen, 1853).
1900: 23°27'08"26 - 0'.'468/, where / is reckoned in years from 1900 (New-
comb, 1895d).
The apparent obliquity (to 0'.'01) was tabulated at intervals of 10 days (5 days
in 1900), for 0h Washington sidereal time from 1855 to 1864, for Washington mean
noon from 1865 to 1881, and for Greenwich mean noon thereafter.
The obliquity actually used in calculating the ephemerides of the Sun, Moon,
and planets in the volumes for 1865-1899 was taken from Hansen and Olufsen
(1853). In 1900, it was taken from Newcomb (1895a).
13.3363 Precession For the general precession in longitude, the expression given
by Peters (1842) 50'.'2411 + 0"0002268 /, where t is reckoned in years from 1800—
was used until 1900, when the value given by Newcomb (1895d) was adopted—
50"2482 + 0:'00022/, where t is reckoned in years from 1900.
The amount of precession (to 0"01 ) since the beginning of the year was tabulated
at intervals of 10 days (5 days in 1900) for & Washington sidereal time from 1855-
1864, for Washington mean noon during 1865-1881, and for Greenwich mean noon
thereafter.
13.3364 Nutation From 1855 to 1899, the term ''equation of the equinoxes in lon-
gitude" was used for the nutation in longitude. The nutations in obliquity used for
the computations relating to the stars were calculated from the formulas given by
Peters (1842). The formulas of Peters were given in the appendix to the volume for
1855; during later years, provision was made for including additional small terms
when required. In the volume for 1900, the nutations of obliquity were taken from
Newcomb (1895a).
From 1865 to 1891. the apparent obliquity and nutation used in the ephemerides
of the Sun, Moon, and planets were taken from Hansen and Olufsen (1853). In 1900.
the obliquity and the long-period nutation were taken from Newcomb (1895a).
The equation of the equinoxes in longitude (to Ol'Ol) and in right ascension (to
0?01) for 1855-1877, and to 0!001 for 1878-1900 were tabulated at intervals of 10
days (5 days in 1900) for 0\ Wishington sidereal time in 1855-1864, for Washington
mean noon in 1865-1881, and for Greenwich mean noon thereafter. In 1900, the
nutation in obliquity (to 0"01) was also explicitly tabulated at intervals of 5 days.
13.3365 Constants The values and authorities used for solar parallax, constant of
aberration, and constant of nutation are as follows:
Solar parallax
1855-1869: 8!'5776 Encke (1824).
638 EXPLANATORY SUPPLEMENT
1870-1899: 8'.'848 Newcomb (1867).
1900: 8"80 Paris (1896). (This value was used for eclipse calculations
from 1896 onward.)
Constant of aberration
1855-1899: 20'.'4451 Struve (1844). (In the ephemeris of the Sun for 1869-
1874, the value 20'.'255 from Hansen and Olufsen (1853)
was used.)
1900: 201'47 Paris (1896).
Constant of nutation
1855-1899: 91'2231 + 0"0009 T, where T is in centuries from 1800.0. Peters (1842).
1900: 9'.'21 at 1900.0 Paris (1896).
13.3366 Semidiameters The values and authorities for the Sun, Moon, and plan-
ets are:
Sun
1855-1899: 16'02" at mean distance (Greenwich observations).
1883-1899: In the calculation of eclipses, 15'59:'78 (Bessel 1830a).
1900: 15'59'.'63 at mean distance (Auwers 1891). This value was used in
calculating eclipses, but in the ephemeris of the Sun, lll5 was added
to it for irradiation.
Moon
1855-1868: Burkhardt's (1812) value increased by 1/500 part.
1869-1900: 0.272274 = 2"5 for irradiation; the irradiation was omitted in the
calculation of eclipses and occupations.
Planets
log distance
1855-
1900:
Mercury
3'.'34
0.00
Venus
8"546
0.00
Mars
21'842
0.25
Jupiter
18:78
0.70
Saturn
8*77
0.95
Uranus
1'.'68
1.30
1882-
1900:
Neptune
1 "28
1.48
13 / HISTORICAL INFORMATION
639
The value for Mercury was taken from Leverrier; the others were determined
by Peirce from observations in 1845 and 1846 with the mural circle at Washington.
The values for Jupiter and Saturn were the polar semidiameters. It was stated in
the 1869 volume that 19'.'19 was erroneously used in 1858-1869 for Jupiter in the
Washington ephemeris. In the volumes for 1869-1900, the equatorial semidiameter
of Jupiter was given as 20"00 at log distance 0.70, and that of Saturn as 9"38 at log
distance 0.95, without authority.
1 3.34 The Nautical Almanac, and The American Ephemeris (1 901 -1 983)
The details given in this section and The Astronomical Almanac for 1901 to 1959
refer specifically to The Nautical Almanac. Differences in the corresponding editions
of The American Ephemeris are specially noted in brackets in which "A.E." refers
only to The American Ephemeris.
From 1960 through 1980, The Astronomical Ephemeris and Nautical Almanac
and The American Ephemeris and Nautical Almanac were completely in agreement,
scientifically. Beginning in 1981, The Astronomical Almanac replaced those two
publications and was printed in a single printing. For the edition for 1968, the IAU
System of Astronomical Constants was introduced. The supplement to the 1968
publications gives the details of the changes to the ephemerides as a result of this
new value of constants.
13.341 Sun The quantities tabulated, generally with argument mean time,
are substantially the same as before: longitude, latitude, right ascension, decli-
nation, and log radius vector; the precision, normally: O'.'l, 0'.'01, 0?01, and 7D; and
(except for the mean longitude and anomaly and the equatorial rectangular coor-
dinates) the interval one day. The mean longitude (to 0?00001 for 1906-1911, and
to 0?0001 since 1912) was given at intervals of 10 days since 1906. [In A.E., the
mean longitude (to 0?0001), at intervals of 5 days, was given since 1934.] Equato-
rial rectangular coordinates (to 7D) was referred to the true equator and equinox
of date, with reductions to those of the beginning of the year, were given at inter-
vals of 12" and until 1930, while similar coordinates at intervals of one day for the
equinox of the beginning of the year were tabulated since 1931, and those for the
standard equinox of 1950.0 (Comrie, 1926) since 1928. (The coordinates for 1928
were given in the volume for 1929.) [In A.E., the interval of 12" was retained until
1950; from 1931 to 1950 the coordinates were referred to the beginning of the year,
and the reductions to the true equinox of date were also tabulated. Coordinates for
1950.0 were tabulated since 1938, at intervals of 12h until 1950, and intervals of one
day from 1951.] Since 1954 [in A.E., since 1953] the coordinates on the "standard"
10-day dates (IAU, 1950) were emphasized by the use of bold type.
640 EXPLANATORY SUPPLEMENT
Longitude, referred to the mean equinox of the beginning of the year, was
included from 1931 to 1937 [in A.E., from 1901 to 1915 and from 1931 to 1951],
Longitude and latitude, which referred to the mean equinox (to 0?00001 ) of 1950.0,
were tabulated at intervals of one day from 1928 to 1937 [not in A.E.]; from 1938
to 1959 [also in A.E.] longitude (to 0"1), latitude (to 0'.'01) were given. [In A.E.,
latitude referred to ecliptic of date given during 1901 1959.] Natural values of the
radius vector (to 7D) were introduced in 1928 [in A.E., in 1938], the logarithmic
value being omitted from both almanacs in 1983.
From 1960 to 1980, the longitude of the Sun was given with respect to the mean
equinox of the beginning of the year. Starting in 1981, the ecliptic longitude was
given for the mean equinox of date. The geocentric rectangular coordinates of the
Sun were omitted in 1981 but reinstated from 1982 onward, and given with respect
to the mean equator and equinox of 1950.0 through 1983.
1901-1980: Newcomb tables (1895a).
1981-1983: Newcomb theory (1895a).
13.342 Moon Values of the longitude and latitude (to 0'.'1), of the parallax (to
0'.'01) [in A.E., (to O'.'l ) until 1912], and of the semidiameter (to O'.'Ol) [in A.E., until
1939] were tabulated at intervals of 12\ Values of the right ascension and declination
(to 0!01 and O'.'l) were given at intervals of one hour, while lunar distances continued
to appear until 1906 [in A.E., until 1911], and examples of their calculation were
given in the issues for 1907-1919 [in A.E., for 1912-1935]. Elements of the mean
equator and orbit have been included throughout. [The angular distance from the
Sun (to O'.l) was given in A.E. from 1937 to 1941, and to l'.O, thereafter.]
In 1981, the hourly ephemeris of the Moon was replaced by a daily ephemeris
in polynomial form for interpolation. The apparent longitude and latitude (each to
0?01) of the Moon were given only daily from 1981 onward.
1901-1922: Hansen (1857) with Newcomb's corrections (1878b) to right ascension.
declination, parallax, and semidiameter from 1901 to 1914, and to
longitude and latitude (before conversion to equatorial coordinates)
from 1915 to 1922. [A.E. for 1912 states that Newcomb's formula for
the correction to Hansen's mean longitude was not, in fact, used for
the years 1883 to 1911. The formula actually used was quoted.]
1923- 1959: Brown (1919). It is to be noted that the values of the longitude (and
correspondingly of the right ascension and declination) for the year
1923 require a small correction of +0'.'08cos(f - /"")-
1959-1983: Brown's theory as given in the Improved Lunar Ephemeris (1954).
13.343 Major Planets Values of the apparent right ascension and declination to
0!01 and O'.'l were tabulated at intervals of one day throughout the period. [In A.E.,
13 / HISTORICAL INFORMATION 54-)
the interval for Uranus and Neptune was 4 days until 1935.] Log distance at the
same interval and to 7D was given until 1934 for Mercury and until 1940 for the
other planets. [In A.E., the intervals until 1915 were 12h for Mercury, one day for
Venus and Mars, 2 days for Jupiter and Saturn and, from 1916, one day for each of
these five; for Uranus and Neptune, it was 4 days from 1901 to 1935.] Natural values
of the distance (to 6D for Mercury to Jupiter, and to 5D Saturn to Neptune) were
given since 1935 for Mercury and since 1941 for the other six planets at intervals
of one day.
Heliocentric longitude and latitude (to 0"1), referred to the mean equinox of
date, and log radius vector (to 7D) were tabulated at intervals of one to four days
until about 1915, and thereafter they were given, in the appendices to various
almanacs between 1915 and 1920, at wider intervals (up to 40 days) for the period
up to 1940 [in A.E., at intervals of one to ten days throughout the period 1901-
1959]. The first two volumes of Planetary Co-ordinates (H.M.N. A.O., 1933 and
1939) contained similar coordinates (to 0P001) or 0P0001, but referred to the mean
equinox of 1950.0, and natural values of the radius vector (to 4D or 5D) for the
period 1920 to 1960, except that no tabulations are given for Mercury and Pluto.
The volumes also contained coordinates of the four outer planets Jupiter, Saturn,
Uranus, and Neptune for 1800 to 1920.
Astrometric right ascensions, declinations, and distances of Pluto have been
included since 1950, but no heliocentric coordinates of this planet have been given.
[In A.E., heliocentric longitude and latitude (to 0"1) and log radius for the period
1960-1983, the right ascensions (to 0!001) and declinations (to 0"01) of the outer
planets were tabulated.]
Mercury and Venus
1901-1980: Newcomb Tables (1895b and 1895c).
1981-1983: Newcomb Theories (1895b and 1895c).
Mars
1901: Leverrier (1861b). [In A.E., tables in manuscript constructed from
elements by Newcomb (1895d).]
1902: Leverrier (1861b). [In A.E., Newcomb (1898).]
1903-1921: Newcomb (1898a).
1922- 1980: Newcomb Tables (1898a) with Rosa's corrections (1917).
1981-1983: Newcomb theory (1898a) with Rosa's corrections (1917).
Jupiter and Satuim
1901-1959: Hill (1895A and 1895b).
642 EXPLANATORY SUPPLEMENT
Uranus
1901-1903: Leverrier (1877a). [In A.E., Newcomb (1865).]
1904-1959: Newcomb (1898b).
1901-1902: Leverrier (1877b). [In A.E., Newcomb (1865).]
1903: Leverrier (1877b). [In A.E., Newcomb (1898c).]
1904-1959: Newcomb (1898c).
Pluto
1950-1959: Bower (1931),
Jupiter, Saturn, Uranus, Neptune, and Pluto
1960-1983: Eckert, Brouwer and Clemence (1951). Geocentric ephemerides only
include perturbations by inner planets (Clemence, 1954).
13.344 Minor Planets Elements and ephemerides of the first four minor planets
were published as in previous years [not in A.E.] from 1901 to 1913; ephemerides
alone were given in 1914 and 1915. Thereafter, no tabulations were given until
1952. Since that year ephemerides have been included to cover, at intervals of one
day, the periods during which transit occurs between sunset and sunrise at (most)
fixed observatories. Although in 1958, the period of tabulation was altered to that
period of time during which the planet is "not within about 40° of the Sun." The
quantities given were apparent right ascension and declination (to 0*01 and 0'.'1)
with corrections "astrometric minus apparent," and distance (to 6D).
Starting in 1981 astrometric positions (to 0s.l) and declination (to l'.'O), respec-
tively, were tabulated at. two-day intervals for Ceres, Pallas, Juno, and Vesta. In
addition, opposition dates, magnitudes, and osculating elements were given for the
brighter minor planets coming to opposition during the year.
The basis of the tabulation has been:
Ceres
1901-1915: Godward (1878).
Pallas
1901-1915: Farley (1856a).
Juno
1901-1915: Hind (1855), with corrections by Downing (1890).
13 / HISTORICAL INFORMATION
Vesta
643
1901-1906: Farley (1856b).
1907-1915: Leveaau (1896).
Ceres, Pallas, Juno, and Vesta
1952-1959: Herget, Clemence, and Hertz (1950).
1960-1971: Herget (1962).
1972-1983: Duncombe (1969).
13.345 Satellites Satellite information is given for Mars, Jupiter, Saturn, Uranus,
Neptune, and Pluto.
13.3451 Mars Diagrams of the apparent orbits and times of elongations to a pre-
cision of , were given for a period of a month on each side of opposition from
1931 to 1980. From 1931 [in A.E. from 1920], tables were added for calculating the
position angle and apparent distance of each satellite from the planet at any time
during the same period. From 1981, times of each eastern elongation for Deimos
and every third eastern elongation for Phobos were given for the entire year. Tables
for calculating the position angle and apparent distance were also extended to cover
the entire year.
1901-1902: Elements by Hall (1878). [In A.E., those of Harshman.]
1903-1915: Elements by Harshman (1894).
1916-1983: Elements by Struve (1911, p. 1073).
13.3452 Jupiter Phenomena and configurations of Satellites I-IV are given from
1901, with considerable variation from time to time in the precision, which is usually
higher for eclipses than for other phenomena. Elongations every twentieth on each
side of Satellite V were given in 1906-1980 [in A.E. 1901-1980]. From 1981, times of
every twentieth eastern elongation have been given for the whole year. Differential
coordinates of Satellites VI and VII have been given since 1931 [in A.E. since 1912].
Sidereal periods of Satellites VIII to XI, and XII have been included since 1953 and
1957. The sidereal period of Satellite XIII was introduced in 1984. Differential
coordinates of Satellites VIII to XIII were introduced in 1981.
Satellites [-TV
1901-1913: Damoiseau (1836) and later extensions. [In A.E., also Woolhouse
(1833).]
1914-1915: Sampson (1910). [In A.E., Damoiseau and extensions and Woolhouse.]
g44 EXPLANATORY SUPPLEMENT
1916: Sampson (1910), with Andoyer's (1915) modifications. [In A.E. until
1930, the configuration are attributed to Pottier's (1896) continuation
of Damoiseau.]
Satellite V
1901-1905
1906-1915
1916-1959
1960-1983:
[In A.E. only, Robertson's elements (unpublished).]
Chon (1897). [In A.E., Robertson.]
Robertson (1924). [In A.E.; 1930-1933, "from Connaissance des Temps
for 1915"; 1934-1959, "from Connaissance des Temps every year be-
ginning with 1919".]
van Woerkom (1950).
Satellites VI- VII
1912-1930: [In A.E. only, Ross (1907a and 1907b).]
1931-1947: Ross (1907a and 1907b).
1948-1983: Bobone (1937a and 1937b).
Satellites VIII-XII
1981-1983: Herget (1968).
Satellite XIII
1981-1983: Aksnes (1978).
13.3453 Saturn Until 1980, diagrams were given for part of each year of the orbits
of Satellites I- VII. The volume also included times of elongations and conjunctions
of Satellites I- VIII to a precision of 06.1. Beginning in 1981, the tables cover the
entire year. Differential coordinates of Satellite IX (Phoebe) were introduced in
1931 [in A.E. in 1909], as well as those of Satellites VII and VIII [in A.E. in I960].
Elements for determining the distance and position angle were added in 1931 [in
A.E. in 1912]. During 1935-1959, The Nautical Almanac [not A.E.] contained, for
about nine months in each opposition, quantities to assist in the calculation of the
phenomena (eclipses, occulations, transits, and shadow transits) of Satellites I- VI.
Satellites I-V: Mimas, Enceladus, Tethys, Dione. and Rhea
1901-1915: Tables in manuscript "prepared by" Newcomb, except the elongations
of Satellites I (Mimas) and III (Tethys), which are from H. Struve
13 / HISTORICAL INFORMATION 045
(1898). [In A.E.; 1901-1903, Hall (1886); 1904, Hall, except the elon-
gations of Satellites I and III (from Struve, 1898); 1905-1913, Struve
(1898); 1914-1915, Struve (1888, 1898, and 1903).]
1916-1930: Struve (1888, 1898, and unpublished corrections).
1931-1935: H. Struve (1888, 1898, and 1903) and G. Struve (1924).
1936-1983: G. Struve (1924 and 1930).
Satellite VI (Titan)
1901-1915:
1916-1930
1936-1937
1938 1983
Newcomb's manuscript tables. [In A.E.: 1901-1904, Hall (1886); 1905-
1913, Struves (1898); 1914-1915, Struve (1888, 1898, and 1903).]
Struve (as for Satellites I-V).
G. Struve (1933). [In A.E., H. Struve, as in 1931.]
G. Struve (1933).
Satellite VII (Hyperion)
1901-1915: Newcomb's manuscript tables. [In A.E.: 1901-1902, Eichelberger (1892);
1903-1913, Struve (1898); 1914-1915, Struve (as for Satellite VI).]
1916-1935: Struve (as for Satellite VI).
1936-1983: Woltjer (1928).
Satellite VIII (Iapetus)
1901-1902: As for Satellite VII. [In A.E., Hall (1885).]
1903-1935: As for Satellite VII.
1936-1937: H. Struve (1888, 1898, and 1903).
1938-1983: G. Struve (1933).
Satellite IX (Phoebe)
1909-1930: [In A.E. only, Ross (1905).]
1931-1983: Ross (1905).
13.3454 Uranus Diagrams of the orbits of Satellites I-IV, and times of el< >ng;it ions
to a precision of 0*1 were given for about nine months in each opposition in 1901
1980. Elements for the calculation of distance and position angle were included in
1927 [in A.E., in 1912]. From 1981, the times of elongations and elements for cal-
culating the distance and position angle have been given for the entire year. The
sidereal period of Satellite V (Miranda) was added in 1953. In 1981, elements for
the rings of Uranus were introduced, as well as times of eastern elongation and
tables for the calculation of the distance and position angle of Satellite V.
646
Satellites I-IV
EXPLANATORY SUPPLEMENT
1901- 1915: Newcomb (1875).
Satellites I and II
1916-1983: Newcomb.
Satellites III and IV
1913-1983: Struve (1913).
Rings
1981-1983: Elliot et al. (1978).
Satellite V
1981-1983: Dunham (1971).
13.3455 Neptune Diagrams of the orbit of Triton, and times of elongations to a
precision of (Kl, were given for about ten months in each opposition in 1901-1980.
Elements for the calculation of distance and position angle were included in 1927
[in A.E., in 1912]. From 1981, times of elongation and tables for the calculation of
distance and position angle covered the whole year. The sidereal period of Nereid
was added in 1953. Times of eastern elongation, and tables for the calculation of
distance and position angle of Nereid were added in 1981.
Triton
1901: Newcomb (1875). [In A.E., Hall (1898).]
1902- 1929: Hall (1898).
1930-1983: Eichelberger and Newton (1926).
Nereid
1981-1982: Rose (1974).
1983: Mignard (1981).
13.3456 Pluto In 1981, times of northern elongation were added for Charon.
1981: Christy and Harrington (1978).
1982-1983: Harrington and Christy (1981).
13 / HISTORICAL INFORMATION
647
13.346 Auxiliary Quantities Auxiliary quantities given are sidereal time, obliq-
uity of the ecliptic, precession, nutation (for both The Nautical Almanac and The
American Ephemeris), and constants.
13.3461 Sidereal time The entries for sidereal time include:
1901-1932: The tabulated values (to 0101) at intervals of one day were based on
Newcomb's value (1895a) for the right ascension of the mean Sun af-
fected by aberration— 18h38m45!836 + 86401841542 T+0!0929 T2, where
T was measured in Julian centuries from 1900 January 0 at 12h UT,
and included the effect of long-period terms only of nutation.
1933-1959: The precision was changed to 01001, and the effect of short-period
terms of nutation was included also.
1960-1983: Tabulated separately for both apparent and mean equinox. Universal
Time of transit of first point of Aries also tabulated.
13.3462 Obliquity of the ecliptic Newcomb's value (1895a) for the mean obliquity
was used throughout: 23°27,08,.'26-46'.,845 7'-0'.'0059r2+0'.'00181 P. [In addition,
A.E. includes values attributed to Hansen and to Peters for the years 1901 to 1915,
also those due to Leverrier for the years 1902 to 1915.]
Values of the true obliquity (to O'.'Ol ) excluding the effect of short-period terms
of nutation were given at intervals of one day until 1930 [in A.E., until 1933; and at
intervals of 5 days from 1934 to 1959]. From 1931 to 1959, only the mean obliquity
for the beginning of the year and the daily nutation in obliquity were given. From
1960 to 1980, the obliquity of the ecliptic was given daily with the Solar ephemeris
and after that it was given with the Besselian day numbers.
13.3463 Precession The values given were based, throughout the period, on New-
comb's determination (1897, p. 73)— 50"2564 + 0'.'0222 7' [A.E. for the years 1901
to 1911 also gave values based on Peters (1842).]
Values of the precession (to O'.'Ol ) from the beginning of the year were tabulated
at intervals of one day [in A.E., 5 days until 1915], while the daily and annual
increments (to O'.'OOOl) were given until 1930 [in A.E., until 1915],
From 1931 [not in A.E.] additional precessional constants, and tables for reduc-
tion of star positions, were included also.]
13.3464 Nutation: The Nautical Almanac The terms used in The Nautical Alma-
nac from 1901 to 1959 were:
1901-1902: The tabulations of long-period and short-period terms (separately)
in both longitude and obliquity were based on the values given by
g48 EXPLANATORY SUPPLEMENT
Newcomb (1895a), as modified by the Paris Conference (1896). Six
long-period terms, and seven short-period, were included, as shown
in Table 13.348.1. The precision was 0"01.
1903-19H: The coefficients of a few of the terms were modified in accordance
with the revised values given by Newcomb (1898d).
1912-1930: An additional term, (8), was included in the day numbers throughout
this period, but only in the tabulation of the nutation from 1918.
1931-1936: Nutation in obliquity and the short-period nutation in longitude were
no longer tabulated explicitly, though the former was available as -B
(the day number). Nutation in right ascension (to 0?001), including
short-period terms, was tabulated.
1937-1959: Tabulations were the same as in former years, but 21 terms (1-4, 6-
10, 14-25) were included. The coefficients used were those given by
Newcomb (1898d) and are quoted in Table 13.348.1.
13.3465 Nutation: The American Ephemeris The terms used in The American
Ephemeris from 1901 to 1983 were:
1901-1911: The tabulations of long-period terms (five-day intervals) and of short-
period terms (one-day intervals) were based on both the values of
Peters (1842) and those of the Paris Conference (1896). The terms
included were 1-3, 5, 6, and 8-15. From 1901 to 1911, the argument
Sun's true longitude was used instead of Sun's mean longitude (L) in
the Peters calculation for terms 3, 5, 8, 6, 10, and 11, and from 1901 to
1907 in the Paris calculations for terms 8, 10, and 11. The coefficients
only differ in a very few cases from those shown in Table 13.348.1, and
the individual discrepancies (usually of a single unit) are not listed.
1912-1936: The tabulations were based solely on the values of the Paris Con-
ference, and after 1915 were all at an interval of one day; the terms
included were 1-4, 6-8, 14, 15, and 17-21.
1937-1959: Tabulation of long-period nutation in longitude, and short-period nu-
tation in both longitude and obliquity, were given at intervals of one
day, and included the terms 1-4, 6-10, and 14-25. The coefficients
are those listed in Table 13.348.1.
1960-1983: The formulas adopted for computing nutation in longitude and obliq-
uity were obtained by retaining all terms for the coefficients as great
as 0!'0002 from the expression developed in Woolard (1953).
13.3466 Constants The following constants were used for the period 1901-1983.
13 / HISTORICAL INFORMATION
Solar parallax
649
1901-1968a: 8"80 Paris (1896, p. 54).
1968-1983: 8'.'79405.
Constant of aberration
1901-1968: 20M7 Paris (1896, p. 54).
1968-1983: 20'.'4958.
Constant of nutation
1901-1936: 9'.'21 Paris (1896, p. 54).
1937-1983: 9"210 + 0'.'0009 T Newcomb (1898d, p. 241).
[A.E. for the years 1901-1911 included tables for the Struve and Peters con-
stants of aberration and nutation as used formerly, as well as those for the above
Paris values.]
An IAU System of Astronomical Constants was adopted in 1964 and introduced
in the 1968 Astronomical Ephemeris and American Ephemeris. The complete list
of constants is given in Section 13.35.
13.347 Ephemeris for Physical Observations This section covers ephemerides for
the Sun, Moon, Mercury, Venus, Mars, Jupiter, and Saturn. In addition, semidiam-
eters are given for the years 1901-1983.
13.3471 Sun An ephemeris for physical observations of the Sun was first published
in The Nautical Almanac for the Year 1907. Previous to that time, observations
were reduced with the aid of tables privately printed by Warren de la Rue. (See the
volumes of the Greenwich photo heliographic observation.) The American Ephem-
eris first included observations of the Sun in 1913.
13.3472 Moon Ephemeris for physical observations of the Moon, calculated by
Marth, appeared in Monthly Notices of the Royal Society during the last quarter of
the nineteenth century. The ephemeris was introduced intoT7i.e Nautical Almanac
in 1907, and was first included in The American Ephemeris in 1913. But the for-
mulas and tables for the calculation of the optical librations, and the times of the
greatest librations, had been included with the ephemeris of the elements of the
mean equator of the Moon, since 1855.
650 EXPLANATORY SUPPLEMENT
13.3473 Mercury and Venus The physical ephemerides of Mercury and Venus were
added to The American Ephemeris in 1882, and included in The Nautical Almanac
in 1907. Previously, only a small table of the versed sine of the illuminated disk
divided by the apparent diameter, for Venus and for Mercury, had been given.
13.3474 Mars and Jupiter Ephemerides for physical observations of Mars, calcu-
lated by Marth, appeared in Monthly Notices of the Royal Astronomical Society
beginning in 1869, and for Jupiter beginning in 1875. They were continued by
Crommelin after Marth's death, and transferred to The Nautical Almanac in 1907.
They were first included in The American Ephemeris in 1913.
For 1968, the ephemeris for "Physical Observations of Mars" was introduced
based on the IAU System of Astronomical Constants.
13.3475 Saturn The physical ephemeris of Saturn first appeared in the ephemeris
in 1960. The stellar magnitude that is included had previously been given in the
ephemeris for the rings.
13.3476 Semidiameters at unit distance: The Nautical Almanac The values given in
The Nautical Almanac for the period 1901-1959 are:
Sun
1901-1959: 961'.'18
1901-1959: 959'.'63
Auwers (1891).
Auwers (1891, p. 367); for eclipses only.
Mercury
1901-1959: 3'.'34
Leverrier (1843).
I V it us
1901-1920: 8'.'40
1921-1959: 8'.'41
Auwers (1891).
Auwers (1894).
Mars
1901-1959: 4'.'68
Hartwig (1879).
Jupiter (equatorial)
1901-1920: 97"36
1921-1959: 98*47
Schur (1896).
Sampson (1910).
13 / HISTORICAL INFORMATION
Jupiter (polar)
651
1901-1920: 91"10
1921-1959: 9l!'91
Schur (1896).
Sampson (1910).
Saturn (equatorial)
1901 1920: 84'.'75
1921-1959: 83'.'33
Meyer (1883).
Struve (1898).
Saturn (polar)
1901-1920: 76:88
1921-1959: 74'.'57
Meyer (1883).
Struve (1898).
Uranus
1901 1930: 34'.'28
1931-1959: 34'.'28
Hind.
Barnard (1896); See (1902); and Wirtz (1912).
Neptune
1901-1959: 36"56
Barnard (1902).
13.3477 Semidiameters at unit distance: The American Ephemeris The values given
in The American Ephemeris for the period 1901-1959 are:
Sun
1901 1902
1903 1959
1901-1959
960'.'78 Auwers (1891); includes 1'.'15 for irradiation.
96l!'50 Harkness (1899); includes l!'15 for irradiation.
959'.'63 Auwers (1891); for eclipses only.
Mercury
1901-1959: 3" 3 I
Leverrier (1843).
Venus
1901-1919: 8:'5S
1920-1959: 8M1
Peirce.
Auwers (1894).
652
Mars
1901-1919: 5705
1920-1959: 4"68
Jupiter (equatorial)
EXPLANATORY SUPPLEMENT
1901-1919:
1920-1959:
100724
98'.'47
Peirce.
Hartwig (1879).
Sampson (1910).
Sampson (1910),
Jupiter (polar)
1901-1919:
1920-1959:
94'.' 12
91791
Peirce.
Sampson (1910).
Saturn (equatorial)
1901-1911: 83:'60
1912-1919: 84788
1920-1959: 83733
Barnard (1902).
Barnard (1902).
Struve (1898).
Saturn (polar)
1901-1919
1912-1919
1920 1959
Uranus
78"16
77747
74757
1901-1919: 33"52
1920-1959: 34728
Neptune
1901-1919: 38"66
1920-1959 : 36756
Peirce.
Barnard (1902).
Struve (1898).
Peirce.
Barnard (1896); See (1902); Wirtz (1912)
Barnard (1902).
Barnard (1902).
13.3478 Semidiameters at unit distance (1960-1983) From 1960 to 1983 the follow-
ing constants were used:
Mercury
3734 Leverrier (1843).
13 / HISTORICAL INFORMATION
Venus
653
Mars
8741 Auwers (1894).
4768 Hartwig (1879).
Jupiter (equatorial)
98"47 Sampson (1910).
Jupiter (polar)
91791
Saturn (equatorial)
83733
Saturn (polar)
74757
Uranus
Sampson(1910).
Struve (1898).
Struve (1898).
Neptune
34728 Bernard (1896); See (1902); and Wirtz (1912).
36756 Bernard (1902).
13.348 Coefficients of Nutation (1834-1959) Table 13.348.1 gives the values
adopted, beginning with each year shown, of the coefficient of sine (Argument)
in the nutation in longitude (A^), and of cosine (Argument) in the nutation in
obliquity (Ae). A blank indicates that the term was not included in the tabulations
for that year. A leader (...) indicates that the coefficient used was the same as in
the preceding entry. The subheading "Ae" is omitted for those terms that do not
occur in the obliquity. Terms numbered 1 to 13 all have periods greater than 100
days, and are known as long-period terms; terms 14 to 25 have periods shorter than
35 days, and are known as short-period terms.
Table 13.348.1
Adopted Values for Coefficients of Nutation
*- CM
< 5
< ;
« cm <
< i
< S
< ;
i |
o m
O 00
03 00
O O
d d
+ +
CO
03
o
d
+
1"
IV
o
cm
o o
CM CM
S
o
CO
m
**
in
id
m
in
Si
d
O
■ d
o
T
+
: +
+
o
m
Is
a
CM
in
CD
m
K
CM
CM
CM
CM
CM
T
T
: 7
T
CO
to
o
0>
o
ft
CO
o>
o
o
o
o
I
d
1
. d
: 1
cm
CO
OD
CO
CJ5
o
o
O
cm
CM
CM
d
d
■ d
t
+
: +
o
CO
m
o
CO
o
o
o
in
CM
o
CM
CM
CM
CM
o
oS
■ ai
OS
d
+
+
: +
+
+
CO c\t
en ir>
CM CM
1 I
<0 to
co co co Fv
CM CM CM i-
. o
|v N S
7 7 T I
SS CD i- CO r-. |v
m cj> o o co co
co co co o) cn o o>
C8
CO I ^
-J
CM
^ + <
<l i
< ;
-1 .
^ -I <
z I
CM
O
CO
•*
CO
CM
■ in
0
CO
|v
-a-
< a
iv
CD
O
o
d
I
in
CM
o
O
CM
5
d
+
CO
o
o
d
+
+
|v
CM T-
O CM
O O
CM
CM
O
CO
in
an
o
o
a
m
o
o
o
d
o
d
I I I
CO
rv
O
+
O lv
CD
CM
O
+
CO |v i- CO CM CO
CD CJ> O O
00 00 O Ol 05
Table 13.348.1, continued
Adopted Values for Coefficients of Nutation
656
1 3.35 System of Constants (1 968-1 983)
EXPLANATORY SUPPLEMENT
13.351 System of Astronomical Constants This system of constants was replaced
for the 1984 edition of the Astronomical Almanac by the IAU (1976) System of
Astronomical Constants given on pages K6 to K7 in the Astronomical Almanac
and in Table 16.2.
Defining constants
Number of ephemeris seconds in one
tropical year (1990)
Gaussian gravitational constant
Primary constants
Astronomical unit
Velocity of light
Equatorial radius of the Earth
Dynamical form-factor for Earth
Geocentric gravitational constant
Mass ratio: Earth/Moon
General precession in longitude per
tropical century (1900)
Constant of nutation
Derived constants
Solar parallax
Light-time for unit distance
Constant of aberration
Flattening factor for Earth
Heliocentric gravitational constant
Mass ratio: Sun/Earth
Mass ratio: Sun/(Earth + Moon)
Mean distance of the Moon
Constant of sine parallax for Moon
Figure of the Earth
Equatorial radius (primary)
Polar radius
Square of eccentricity
Reduction from geodetic latitude 0
to geocentric latitude 0'
Radius vector
31556925.9747
0.017202098950000
= 3548'.' 1876069651
149600 x 106 m
299792. 5 x 103 m/sec
6378160 m
0.0010827
398.603 x 109m3s-2
81.30
5025!'C4
9!210
8"794
499s012
20"496
1/298.25
132718 x 1015
332958
328912
384400 x 103m
3422'.'451
3 —2
nr s
6378160 m
a(l -/) = 6356774. 7 m
0.00669454
0'-0 = -ll'32:'743Osin20
+ 1 1633 sin 40 - 0'.'0026 sin 60
p a a(0. 998327073
+ 0.001676438 cos 20
-0. 0000035 19 cos 40
+ 0.000000008 cos 60)
13 / HISTORICAL INFORMATION „„
bo/
Figure of the Earth, continued
One degree of latitude (m) 1 1 133. 35 - 559. 84cos 20
+ 1.17 cos 40
(0 = mid-latitude of arc)
One degree of longitude (m) 1 1413. 28 cos 0 - 93. 51 cos 30
+ 0. 12cos50
The complete system of astronomical constants is given in Supplement to the
American Ephemeris 1968 (pp. 4s-7s).
13.352 Old Constants The printed ephemeris of the Sun and inner planets are
based on the following values of the constants that were in use prior to the intro-
duction of the IAU (1964) System.
Velocity of light 299860 x 1 03 m/sec
Equatorial radius of the Earth 6378388m
Constant of aberration 20'.'47
Flattening factor of the Earth 1/297
Polar radius of the Earth a( \ — /) = 635691 1. 946 m
Solar parallax 8'.'80
Light-time for unit distance 498'.'38
Constant of aberration 20'.'47
Mass ratio:
Sun/(Earth+Moon) 329390
Earth/Moon (planetary theory) 81.45
13.4 REFERENCES
Adams, J.C. (1849). "On an Important Error in Bouvard's Tables of Saturn" Mem. R.A.S.
17, 1-2.
Adams, J.C. (1853). "On New Tables of the Moon's Parallax" Nautical Almanac for 1856
Appendix, 35-43.
Adams, J.C. (1853a). "On the Corrections to be Applied to Burckhardt's and Plana's
Parallax of the Moon, Expressed in Terms of the Mean Arguments" M.N. R.A.S. 13,
262-266.
Adams, J.C. (1877). "Continuation of Tables I and III of Damoiseau's Tables of Jupiter
Satellites" Nautical Almanac for 1881 Appendix, 15-23.
Airy, G.B. (1832). "On an Inequality of Long Period in the Motions of the Earth and
Venus" Phil. Trans. Roy. Soc. 67-124.
Airy, G.B. (1835). "On the Calculation of the Perturbations of the Small Planets and the
Comets of Short Period" Nautical Almanac for 1837 Appendix, 149-171.
Airy, G.B. (1845). Reduction of the Observations of the Planets (1750-1830) London.
Printed by Palmer and Clayton, and sold by J. Murray.
ggg EXPLANATORY SUPPLEMENT
Airy, G.B. (1848). Reduction of the Observations of the Moon (1750-1830) 2 volumes
London. Printed by Palmer and Clayton, and sold by J. Murray.
Airy, G.B. (1849). "Corrections of the Elements of the Moon's Orbit, Divided from the
Lunar Observations Made at the Royal Observatory of Greenwich, from 1750 to 1830"
Mem. R.A.S. 17, 21-57.
Airy, G.B. (1855). Observations Made at the Royal Observatory, Greenwich, in the Year
1853 (Her Majesty's Stationary Office, London).
Aksnes, K. (1978). "The motion of Jupiter XIII (Leda), 1974-2000" Astron. J. 83, 1249-
1256.
Andoyer, H. (1915). "Sur le Calcul des Ephemerides des Quatres Anciens Satellites de
Jupiter" Bull Astr. 32, 177-224.
Auwers, A. (1891). "Der Sonnendurchmesser und der Venusdurchmesser nach den Beobach-
tungen an den Heliometern der Deutschen Venus-Expeditionen" Ast. Nach. 128, 361-
376.
Auwers, A. (1894). "Bemerkung zu den Mittheilungen von Resultaten der Deutschen
Venus-Expeditionen in Nr. 3066 und 3068 der A.N." Ast. Nach. 134, 359-362.
Baily, F. (1825). "On the Construction and Use of Some New Tables for Determining the
Apparent Places of Nearly 3000 Principal Fixed Stars" Mem R.A.S. 2, Appendix,
i-ccxxviii.
Bessel, F.W. (1828). "Uber den Gegenwartigen Zustand Unserer Kenntniss der Sonnen-
bewegung und die Mittel zu Hirer Verbesserung" Ast. Nach. 6, 261-276 and 293-302.
Bessel, F.W. (1830a). Tabulae Regiomontanae Konigsberg.
Bessel, F.W. (1830b). "Vorlaufige Nachricht von Einem auf er Kouigsberger Sternwarte
Befindlichen Grossen Heliometer" Ast. Nach. 8, 397-426.
Bessel, F.W. (1875a). "Untersuchungen liber den Planeten Saturn, Seinen Ring und Seinen
Vierten Trabanten" Abhandlungenl, 150-159.
Bessel, F.W. (1875c). "Ueber die Neigung der Ebene des Saturnsringes" Abhandlungen I,
319-321.
van Biesbroeck, G. (1957). "The mass of Neptune from a New Orbit of its Second Satellite
Nereid" Astron. J. 62, 272-274.
Bobone, J. (1937a). "Tables del VI (sexto) Satellite de Jupiter" Ast. Nach. 262, 321-346.
Bobone, J. (1937b). "Tables del VII (septimo) Satellite de Jupiter" Ast. Nach. 263, 401-
412.
Bouvard, A. (1821). Tables Astronomiques, ...Contenant les Tables de Jupiter, de Saturne
et d'Uranus (Bachelier et Huzard, Paris).
Bower, E.C. (1931). "On the Orbit and Mass of Pluto with an Ephemeris for 1931-1932"
Lick Obs. Bull. 15, 171-178.
Breen, H. (1848). "Corrections of Lindenau's Elements of the Orbit of Venus Deduced from
the Greenwich Planetary Observations, from 1750-1830" Mem. R.A.S. 18, 95-154.
Breen, H. (1851). "On the Corrections of Lindenau's Elements of Mars" Mem. R.A.S. 20,
137-168.
Brown, E.W. (1919). Tables of the Motion of the Moon (Yale University Press, New Haven).
Burckhardt, J.C. (1812). Tables de la Lune in Tables Astronomiques publiees par le Bureau
des Longitudes de France (Mme. Ve Courcier, Paris).
Burg, J.T. (1806). Tables de la Lune in Tables Astronomiques publiees par le Bureau des
Longitudes de France (Mme. Ve Courcier, Paris).
Carlini, F. (1810). Exposizione di un nuovo metodo di construire le Tavole Astronomiche
in Applicato alle Tavole de Sole (Reale Stamperia, Milan).
13 / HISTORICAL INFORMATION 559
Carlini, F. (1832). "Nuove Tavole de' Moti Apparenti del Sole" Effemeridi...di Milano per
VAnno 1833 Appendix, 1-104.
Christy, J.W. and Harrington, R.J. (1978). "The Satellite of Pluto" Astron. J. 83, 1005-
1008.
Clemence, G.M. (1954). "Perturbations of the Five Outer Planets by the Four Inner Ones"
Astron Papers for the American Ephemeris XIH, V
Cohn, F. (1897). "Bestimmung der Bahnelemente des V. Jupiters Mondes" Ast Nach
142, 289-338.
Comrie, L.J. (1926). "The Use of a Standard Equinox in Astronomy" M.N.R.A.S 86
618-631.
Comrie, L.J. (1934). "Phenomena of Saturn's Satellites" Mem. B.A.A. 30, 97-106.
Crawford, R.T. (1938). "The Tenth and Eleventh Satellites of Jupiter" P.A.S.P. 50 344-
347.
de Damoiseau, M.C.T. (1824). Tables de la Lune formees par la seule theorie de Vattraction
(Bachelier, Paris).
de Damoiseau, M.C.T. (1836). Tables ecliptiques des satellites de Jupiter (Bachelier, Paris).
Delambre, J.B.J. (1806). Tables du Soleil (Bachelier, Paris).
Delambre, J.B.J. (1817). Tables ecliptiques des satellites de Jupiter (Bachelier, Paris).
Downing, A.M.W. (1890). "Corrections to the Elements of the Orbit of Juno" M.N.R.A.S.
50, 487-497.
Duncombe, R.L. (1969). "Heliocentric Coordinates of Ceres, Pallas, Juno, Vesta 1928-
2000" Astronomical Papers for the American Ephemeris XX, II.
Dunham, D.W. (1971). "The Motions of the Satellites of Uranus" Dissertation. (Yale
University, New Haven, CT).
Eckert, W.J., Brouwer, D., and Clemence, G.M. (1951). "Coordinates of the Five Outer
Planets" Astronomical Papers for the American Ephemeris XII.
Eichelberger, W.S. (1892). "The Orbit of Hyperion" Astron. J. II, 145 157.
Eichelberger, W.S. and Newton, A. (1926). "The Orbit of Neptune's Satellite and the
Pole of Neptune's Equator" Astronomical Papers for the American Ephemeris IV,
275-337.
Elliot, J.L., French, R.G., Frogel, J. A., Elias, J.H., Mink, D.J., and Liller, W. (1981).
"Orbits of Nine Uranian Rings" Astron. J. 86, 444-455.
Encke, J.F. (1824). Der Venusdurchgang von 1769 (Becker, Gotha).
Encke, J.F. (1852a). "Uber eine Neue Methode der Berechnung der Planetenstorungen"
Astr. Nach. 33, 377-398.
Encke, J.F. (1852b). "Zusatz zu dem Aufsatze: Neue Methode zur Berechnung der Speciellen
Storungen in Nr. 791 und 792 der Astron. Nachrichten" Ast. Nach. 34, 349-360.
Farley, R. (1856a). "Correction of the Elements of Pallas" Nautical Almanac for 1860
570-572.
Farley, R. (1856b). "Correction of the Elements of Vesta" Nautical Almanac for 1860
573-575.
Farely, R. (1856c). "Elements of Astraea" Nautical Almanac for 1860 576.
Godeard, W. (1878). "On the Correction of the Elements of Ceres" M.N.R.A.S. 38, 119-
122.
Grosch, H.R.J. (1948). "The Orbit of the Eight Satellites of Jupiter" Astron. J. 53, 180-
187.
Hall, A. (1878). "Observations and Orbits of the Satellites of Mars" Washington Obser-
vations for 1875 Appendix, 1-46.
660
EXPLANATORY SUPPLEMENT
Hall, A. (1885). "The Orbit of Iapetus, the Outer Satellite of Saturn" Washington Obser-
vations for 1882 Appendix I, 1-82.
Hall, A. (1886). "The Six Inner Satellites of Saturn" Washington Observations for 1883
Appendix I, 1-74.
Hall, A. (1898). "The Orbit of the Satellite of Neptune" Astron. J. 19, 65-66.
Halley, E. (1749). Tabulae Astronomicae, accedunt de usu tabularum praecepta (Gulielmum
Innys, London).
Hansen, P.A. (1847). "Aufzug aus Einem Briefe des Herrn Professors Hansen an den
Herausgeber" Ast. Nach. 25, 325 332.
Hansen, P.A. (1857). Tables de la Lune construites d'apress le principe newtonien de la
gravitation universelle (G.E. Eyre et G. Spottiswoode, London).
Hansen, P.A. and Olufsen, C.F.R. (1853). Tables du Soleil (B. Luno, Copenhagen).
Harkness, W. (1899). [The following is taken from the preface to The American Ephem-
eris for 1903: "on a recent investigation by my self, which rests upon 35, 842 merid-
ian observations made at Greenwich, Paris, Washington, Konigsberg, Madras, Milan,
Dorpat, Modena, and Seeberg."]
Harrington, R.J. and Christy, J.W. (1981). "The Satellite of Pluto III" Astron. J. 86,
442-443.
Harris, D.L. [Recent (unpublished) work on the satellites of Uranus quoted by Kuiper
(1956, p. 1632).]
Harshman, W.S. (1894). "The Orbit of Deimos" Astron. J. 14, 145-148. [Harshman's
elements of Phobos "(communicated by Prof. Harkness)" are attached (in manuscript)
to a copy of the above paper. It is believed that they were never published. See also
Publ. U.S.N.O., 6, Appendix I, B21 (1911).]
Herget, P (1938). "Jupiter XI" H.A.C. 463.
Herget, P., Clemence, G.M., and Hertz, H.G. (1950). "Rectangular Coordinates of Ceres,
Pallas, Juno, Vesta, 1920-1960" Astronomical Papers for the American Ephemeris II,
521-587.
Herget, P. (1966). "Rectangular Coordinates of Ceres, Pallas, Juno Vesta 1960-1980"
Astronomical Papers for the American Ephemeris XVI.
Herget, P. (1968). "Outer Satellites of Jupiter" Astron. J. 73, 737-742.
H.M. Nautical Almanac Office (1933). Planetary Co-ordinates for the Years 1800-1940
Referred to the Equinox of 1950.0 London.
H.M. Nautical Almanac Office (1939). Planetary Co-ordinates for the Years 1940-1960
Referred to the Equinox of 1950.0 London.
Herrick, S. (1952). "Jupiter IXX and Jupiter XII" P.A.S.P. 64, 237-241.
Hill, G.W. (1890). "A new theory of Jupiter and Saturn" Astronomical Papers for the
American Ephemeris IV, 1-577.
Hill, G.W. (1895a). "Tables of Jupiter Constructed in Accordance with the Methods of
Hansen" Astronomical Papers for the American Ephemeris VII, 1-144.
Hill, G.W. (1895b). "Tables of Saturn Constructed in Accordance with the Methods of
Hansen" A.P.A.E. 7, 145-285.
Hind, J.R. (1855). "Revised Elements of Juno" Nautical Almanac for 1859 562-564.
International Astronomical Union (1950). Trans. IAU7, 65 and 83.
Kendall, E.O. (1877). [An extension for the year 1880 of Damoiseau's tables of the satellites
of Jupiter (probably never published).]
Kowalski, M. (1855). Recherches sur les mouvements de Neptune Kasan
Kuiper, G.P. (1949). "The Fifth Satellite of Uranus" P.A.S.P. 61, 129.
13 / HISTORICAL INFORMATION gg^
Kuiper, G.P. (1956). "On the Origin of the Satellites and the Trojans" Vistas in Astronomy
2, 1631-1666. y
de Lalande, J.J. (1771). Astronomie 2nd ed. (Desaint, Paris),
de Lalande, J.J. (1792). Astronomie 3rd ed. (Desaint, Paris).
de Laplace, PS. (1787). "Sur les Inegalitie Seculares des Planets" Memoires de I'Academie
des Sciences de Paris 1784, pp. 1-50.
de Laplace, P.S. (1788). "Theorie de Jupiter et de Saturne" Memoires de I'Academie des
Sciences de Paris 1785, pp. 33-160.
Leveau, G. (1896). "Tables du Mouvement de Vesta, Fondees sur la Comparison de la
Theorie avec les Observations" Ann. de I'Obs. de Paris 22, A.1-A.317.
Leverrier, U.J.J. (1845). "Theorie du Movement de Mercure" Connaissance des Temps for
1848 Additions, 1-165.
Leverrier, U.J.J. (1846). "Researches sur les Mouvements de la Planete Herschel (dite
Uranus)" Connaissane des Temps for 1849 Additions, 1-254.
Leverrier, U.J.J. (1858). "Theorie et Tables du Mouvement Apparent du Soleil" Ann. de
I'Obs. de Paris 4, 1-263 and 1-121.
Leverrier, U.J.J. (1859). "Theorie et Tables du Movement de Mercure" Ann. de I'Obs. de
Paris 5, 1-195.
Leverrier, U.J.J. (1861a). "Theorie et Tables du Movement de Venus" Ann. de I'Obs. de
Paris 6, 1-184.
Leverrier, U.J.J. (1861b). "Theorie et Tables du Mouvement de Mars" Ann. de I'Obs. de
Paris 6, 185-435.
Leverrier, U.J.J. (1876a). "Tables de Jupiter" Ann. de I'Obs. de Paris 12, 1-76 and 1-178.
Leverrier, U.J.J. (1876b). "Tables de Saturne" Ann. de I'Obs. de Paris 12, A.1-A.80 and
A.1-A.286.
Leverrier, U.J.J. (1877a). "Tables du Mouvement. d'Uranus, Fondees sur la Comparison
de la Theorie avec les Observations" Ann. de I'Obs. de Paris 14, A.1-A.92 and A.l-
A.163.
Leverrier, U.J.J. (1877b). "Tables du Mouvement de Neptune, Fondees sur la Comparison
de la Theorie avec les Observations" Ann. de I'Obs. de Paris 14, 1-70 and 1-96.
Levin, A.E. (1931, 1934). "Mutual Eclipses and Occultations of Jupiter's Satellites" J.B.A.A.
42, 6-14 (1931). Mem. B.A.A. 30, 149-183 (1934).
Lindenau, B. (1810). Tabulae Veneris novae et correctae (Becker, Gotha).
Lindenau, B. (1811). Tabulae Martis novae et correctae (Schoeniana, Eisenberg).
Lindenau, B. (1813). Investigation nova orbitae a Mercurio circa Solem descriptae (Becker,
Gotha).
Longstreth, M.F. (1853). "On the Accuracy of the Tabular Longitude of the Moon, to
be Obtained by the Construction of New Lunar Tables" Trans. Amer. Phil. Soc. 10,
225-232.
Marth, A. Marth published a long series of ephemerides for physical observations of the
planets and for observations of the satellites. The first and last of each series are alone
quoted here.
Marth, A. (1870-1894). "Ephemeris of the Satellites of Uranus" M.N.R.A.S. 30, 150-54,
171.
Marth, A. (1873 1895). "Ephemeris of the Satellites of Saturn" M.N.R.A.S. 33, 513-55,
164.
Marth, A. (1878-1891). "Ephemeris of the Satellite of Neptune" M.N.R.A.S. 38, 475-51,
563.
662
EXPLANATORY SUPPLEMENT
Marth, A. (1883-1896). "Data for Finding the Positions of the Satellites of Mars" M.N.R.A.S.
44, 29-56, 435.
Marth, A. (1891-1896). "Ephemeris for Computing the Positions of the Satellites of
Jupiter" M.N.R.A.S. 51, 505-56, 534.
Mason, C. (1780). Lunar Tables London, (no publisher given in book)
Mason, C. (1787). Mayer's Lunar Tables Improved London. Printed by William Richardson
and sold by P. Emsly.
Mayer, T. (1770). Tabulae Motuum Solis et Lunae novae et correetae; puibus accedit
Methodus Longitudinum Promota (Gulielmiet Johannis Richardson, London).
Mignard, F. (1981). "The Mean Elements of Nereid" Astron. J. 86, 1728-1729.
Newcomb, S. (1865). An Investigation of the Orbit of Neptune, with General Tables of its
Motion (Smithsonian, Washington, DC).
Newcomb, S. (1867). "Investigation of the Distance of the Sun and of the Elements Which
Depend upon it" Washington Observations for 1865 Appendix II, 1-29.
Newcomb, S. (1873). An Investigation of the Orbit of Uranus, with General Tables of its
Motion (Smithsonian, Washington, DC).
Newcomb, S. (1875). "The Uranian and Neptunian Systems" Washington Observations
for 1873 Appendix I, 1-74.
Newcomb, S. (1878a). "Researches on the Motion of the Moon" Washington Observations
for 1875 Appendix II, 1-280.
Newcomb, S. (1878b). Corrections to Hansen's Tables of the Moon, Prepared and Printed
for the Use of the The American Ephemeris and Nautical Almanac (U.S. Government
Printing Office, Washington, DC).
Newcomb, S. (1895a). "Tables of the Mot ion of the Earth on its Axis and Around the Sun"
Astronomical Papers for the Ephemeris VI, 1-169. [Usually referred to as Tables of
the Sun.]
Newcomb, S. (1895b). "Tables of the Heliocentric Motion of Mercury" Astronomical Papers
for the American Ephemeris VI, 171-270.
Newcomb, S. (1895c). "Tables of the Heliocentric Motion of Venus" Astronomical Papers
for the American Ephemeris VI, 271-382.
Newcomb, S. (1895d). "The Elements of the Four Inner Planets and the Fundamental
Constants of Astronomy" Supplement to the American Ephemeris for 1897 1-202.
Newcomb, S. (1897). "A New Determination of the Precessional Constant" Astronomical
Papers for the American Ephemeris VIII, 1-76.
Newcomb, S. (1898a). "Tables of the Heliocentric Motion of Mars" Astronomical Papers
for the American Ephemeris VI, 383-586.
Newcomb, S. (1898b). "Tables of the Heliocentric Motion of Neptune" Astronomical Papers
for the American Ephemeris VII, 287-416.
Newcomb, S. (1898c). "Tables of the Heliocentric Motion of Neptune" Astronomical Papers
for the American Ephemeris VII, 419-471.
Newcomb, S. (I898d). "Sur les Formules de Nutation Bases sur les Decisions de la Con-
ference de 1896" Bull. Astr. 15, 241-246.
Nicholson, S.B. (1944). "Orbit of the Ninth Satellite of Jupiter" Ap. J. 100, 57-62.
Paris (1896). Conference Internationale des Etoiles Fondamentales: Process- Vebaux Paris.
Peirce, B. (1848a). "Perturbations of Uranus; Investigations into the Action of Neptune
upon Uranus" Proc. Amer. Acad. Sci (Boston) 1, 144-149 and 332-342.
Peirce, B. (1848b). "Uber die Storungen des Neptune's Ast. Nach." 27, 215-218.
Peirce, B. (1853). Tables of the Moon 2nd ed. 1865, (Bureau of Navigation. Washington,
DC).
13 / HISTORICAL INFORMATION
663
Peters, C.A.F. (1842). Numerus Constants Nutationis ex ascensionibus rectis stellae po-
laris in specula dorpatensi 1822-1838 observatis deductus; adjecta est disquisitio the-
oretica de formula nutationis St. Petersburg.
Plana, J.B. (1832). Theorie du mouvement de la Lune 3 volumes, (Imprimerie Rovale
Turin).
Pottier, L. (1896). "Addition aux Tables Ecliptiques des Satellites de Jupiter de Damoi-
seau" Bull. Astr. 13, 67-79 and 107-112.
Robertson, J. (1924). "Orbit of the Fifth Satellite of Jupiter" Ap. J. 35, 190-193.
Rose, L.E. (1974). "Orbit of Nereid and the Mass of Neptune" Astron. J. 79, 489-490.
Ross, F.E. (1905). "Investigations of the Orbit of Phoebe" Harvard Annals 53, 101-142.
Ross, F.E. (1907a). "Semi-definitive Elements of Jupiter's Sixth satellite" Lick Obs Bull
4, 110-112.
Ross, F.E. (1907b). "New Elements of Jupiter's Seventh satellite" Ast. 174, 359-362.
Ross, F.E. (1917). "New Elements of Mars and Tables for Correcting the Heliocentric Po-
sitions Derived from Astronomical Papers, Volume VI, Part IV" Astronomical Papers
for the American Ephemeris IX, 251-274.
Runkle, J.D. (1855). "New Tables for Determining the Values of the Coefficients, in the
Perturbative Function of Planetary Motion, Which Depend upon the Ratio of the
Mean Distances" Smithsonian Contribution to Knowledge IX, Appl (Smithsonian,
Washington, DC).
Sampson, R.A. (1910). Tables of the Four Great Satellites of Jupiter (William Wesley &
Son, London).
Schubert. E. (1854). "On the Correction for the Elements of (1) Ceres" Astron. J. 3,
153-159 and 162-165.
Seidelmann, P.K. (1979). "The Ephemerides; Past, Present and Future" Dynamics of the
Solar System R.L. Duncombe, ed. pp. 99-114.
Struve, F.G.W. (1844). "Sur le Coefficient Constant dans PAberration des Etoiles Fixes,
Deduits des Observations Qui ont ete Executees a l'Observatoire de Pulkova par
l'lnstrument des Passages de Repsold, Etabli dans le Premier Verical" Mem. de I 'Acad
Imp. Sci. St. Petersburg 6th series, 5, 229-285.
Struve, G. (1924-1933). Neue Untersuchungen im Saturnsystem. Veroffentlichungen der
Universitatssternwarte zu: Berlin- Babelsberg 6, parts 1, 4, 5.
Struve, G. (1924). Part 1. "Die Bahn von Rhea" 1-16.
Struve, G. (1930). Part 4. "Die Systeme Mimas-Tethys und Enceladus-Dione" 1-61.
Struve, G. (1933). Part 5. "Die Beobachtungen der ausseren Trabanten und die Bah-
nen von Titan und Japetus" 1-44.
Struve, H. (1888). "Beobachtungen der Saturnstrabanten" Supplement 1 aux Observations
de Poulkova, 1-132.
Struve, H. (1903). "Neue Bestimmung der Libration Mimas-Tethys" Ast. Nach. 162, 325-
344.
Struve, H. (1911). Uber die Lage der Marsachse und die Konstanten im Mars System
Sitzungsberichte der Koniglich Preussischen Akademie der Wissenschaften fiir 1911
1056-1083.
Struve, H. (1913). Bahnen der Uranustrabanten. Abteilung 1: Oberon und Titania. Ab-
handlungen der Koniglich Preussischen Akademie der Wissenschaften.
Taylor, S.W. (1951). "On the Shadow of Saturn on Its Rings" Astron. J. 55, 229-230.
Todd, D.P. (1876). A Continuation of de Damoiseau's Tables of the Satellites of Jupiter
to the Year 1900 (Bureau of Navigation, Washington, DC).
gg4 EXPLANATORY SUPPLEMENT
Vince S. (1797-1808). A Complete System of Astronomy 3 volumes. 2d ed., 1814-1825
(University of Cambridge, Cambridge).
Walker, S.C. (1848). Report of the United States Coast and Geodetic Survey for 1848.
Wargentin, P.W. (1746). "Tabulae procalculandis eclipsibus satellitum Jovis" Acta Soc.
reg. sci. Upsaliensis ad annum 17^1. Stockholm.
Wilson, R.H. (jr.) (1939). "Revised Orbit and Ephemeris for Jupiter X" P.A.S.P. 51,
241-242.
Winlock, J. (1864). Tables of Mercury, for the Use of The American Ephemeris and Nau-
tical Almanac (Bureau of Navigation, Washington, DC).
van Woerkom, A.J.J. (1950). "The Motion of Jupiter's Fifth Satellite, 1892-1949" Astro-
nomical Papers for the American Ephemeris XIII, 1-77.
Woolard, E.W. (1953). "Theory of the Rotation of the Earth Around its Center of Mass"
Astronomical Papers for the American Ephemeris XV, pt 1.
Woolhouse, W.S.B. (1833). "New Tables for Computing the Occupations of Jupiter's
Satellites by Jupiter, the Transits of the Satellites and Their Shadows over the Disc
of the Planet, and the Positions of the Satellites with Respect to Jupiter at any Time"
Nautical Almanac for 1835 Appendix, 1-39.
Zadunaisky, P.E. (1954). "A Determination of New Elements of the Orbit of Phoebe,
Ninth Satellite of Saturn" Astron. J. 59, 1-6.
13 / HISTORICAL INFORMATION ggg
The reports and recommendations of Commission 4 of the International Astronomical
Union have been published as follows:
Transactions IAU
MssenriDiy
I, 159, 207; 1923.
Rome
1922
II, 18-19, 178, 229; 1926
VIUUUI lUgtr, ILIlglrlllQ
1925
III, 18, 224, 300; 1929
T iP>if*i ph
UC1UCU
IV, 20, 222, 282; 1933.
namnrintro i\T A
v t i 1 1 M 'I J,U.gt!, ivl/l
1 mi
19.3 2
V, 29-33, 281-288, 369-371; 1936
1 ri-j -
VI, 20 25, 336, 355-363; 1939
OLUvlVllUlJll
i mo
VII, 61, 75-83; 1950
Z ii rich
iy4o
VIII, 66-68, 80-102; 1954
rtooie
IX, 80-91; 1957
Dublin
1955
X. 72. 85-99; 1960
iviuacow
1 nco
iy5o
XI A 1-8" 1962 B Ifi4-1fi7 441-4R9- 1QR9
J * *•*) A. XS\J A* \ It. XU"X I'll. 1 1 J 1 I J ^ ^ 1 jUi
Berkeley
1961
XII A 1-10- 1965 B 101-105 .W^-fi?^ IQfifi
-nd.muu.rg
XIII A. 1-9- 1967 B 47-53 178-182- 1968
I 1 i 1 ! . 1 H
lab i
XIV, A, 1-9; 1970, B, 79-85, 198-199; 1977
Brighton
1970
XV, A. 1 10; 1973, B, 69-72; 1974
Sydney
1973
XVI, Al, 1-7; 1976, B, 31, 49-67; 1977
Grenoble
1976
XVII, Al, 1-6; 1979, B, 63-83; 1980
Montreal
1979
XVIII, A, 1 13; 1982, B, 67-72; 1983
Patras
1982
XLX, A, 1-6; 1985, B, 93-96; 1986
New Delhi
1985
XX, A, 16; 1988, B, 105-108; 1989
Baltimore
1988
Protocols of the Procedings of the International Conference held at Washington for
the purpose of fixing a Meridian and a Universal Day. October 1884. Washington D.C.,
1884.
Process- Verbaux of the Conference Internationale des Etoiles Fondamentales de 1896.
Paris, Bureau des Longitudes, 1896.
Congress International des Ephemerides Astronomiques tenu aTObservatoire de Paris
du 23 au 26 Octobre 1911. Paris Bureau des Longitudes, 1912. A full account, with English
translations of the resolutions, is given in M.N.R.A.S., 72, 342-345, 1912.
Colloque International sur les Constantes Fondamentales de l'Astronornie. Observa-
toire de Paris, 27 Mars-ier Avril 1950. Colloques International^ du Centre National de la
Recherche Scientifique, 25, 1 131, Paris, 1950. The proceedings and recommendations are
also available in Bull. Astr. 15, parts 3-4, 163-292, 1950.
Proceedings of IAU Colloquim No. 9 (Heidelburg, 1970) on "The IAU System of
Astronomical Constants," Celestial Mechanics 4, no. 2, 128-280, 1971.
Duncombe, R. L., Fricke, W., Seidelmann, P. K., and Wilkins, G. A. (1977).
Proceedings of the Sixteenth General Assembly, Grenoble 1976, p. 56.
CHAPTER
Related Publications
compiled by M.R. Lukac
14.1 CURRENT PUBLICATIONS
14.1 1 Joint Publications of the Royal Greenwich Observatory and the United
States Naval Observatory
These publications are available from Her Majesty's Stationery Office and from the
Superintendent of Documents, U.S. Government Printing Office, Washington, DC,
20402.
In the United States, the Sight Reduction Tables may be obtained from the De-
fense Mapping Agency, which includes the former Hydrographic Department. The
Defense Mapping Agency's distribution center address is: DMA Combat. Support
Services, Washington, DC, 20315-0010.
The Astronomical Almanac contains the ephemerides of the Sun, Moon, planets
and their natural satellites, as well as data on eclipses and other astronomical
phenomena. It includes information concerning timescales and coordinate systems;
data on stars and stellar systems, including bright stars, standards for photometry,
radial velocity, and radio sources; and lists of bright galaxies, X-ray sources, variable
stars, quasars, and pulsars. It also contains a list of the observatories of the world,
information concerning the calendar, and the Astronomical System of Constants.
The Nautical Almanac contains ephemerides at an interval of one hour and
auxiliary astronomical data for marine navigation.
The Air Almanac contains ephemerides at an interval of ten minutes and aux-
iliary astronomical data for air navigation.
Astronomical Phenomena contains extracts from The Astronomical Almanac
and is published annually in advance of the main volume. It contains the dates and
667
g68 EXPLANATORY SUPPLEMENT
times of planetary and lunar phenomena and other astronomical data of general
interest.
Planetary and Lunar Coordinates, 1984-2000 provides low-precision astronom-
ical data for use in advance of the annual ephemerides and for other purposes. It
contains heliocentric, geocentric, spherical, and rectangular coordinates of the Sun,
Moon, and planets; eclipse data; and auxiliary data, such as orbital elements and
precessional constants.
Sight Reduction Tables for Air Navigation, (Pub. No. 249, reproduced as (British)
Air Publication, A. P. 3270). Volume 1 contains the altitude to V and the azimuth
to 1° for the seven most suitable stars for navigation, for each degree of latitude
and for each degree of local sidereal time. Volumes 2 and 3 give similar data for
each degree of declination to 29° and for each degree of hour angle. Tabulations
extend to depressions of at least 5° below the horizon.
Sight Reduction Tables for Marine Navigation, (Pub. No. 229, reproduced as
(British) Hydrographic Department, N.P. 401, 1971 onward). These tables give alti-
tude to O'.l, with variations for declination and azimuth to 0?1, with arguments for
latitude, hour angle, and declination, all at 1° intervals. They provide all solutions
of the spherical triangle, given two sides and the included angle, to find a third side
and adjacent angle.
14.12 Other Publications of the United States Naval Observatory
Almanac for Computers contains short mathematical series that are used to repre-
sent the positions of the Sun, Moon, and planets for efficient evaluation with small
computers or programmable calculators. Data for both astronomical and naviga-
tional applications are included. Discontinued.
Astronomical Papers of The American Ephemeris are issued irregularly and
contain reports of research in celestial mechanics with particular relevance to ephem-
erides.
United States Naval Observatory Circulars are issued irregularly to disseminate
astronomical data concerning ephemerides or astronomical phenomena.
The Multi-year Interactive Computer Almanac (MICA) is an integrated package
of software and astronomical data on a floppy diskette. MICA will produce to full
precision most of the data in The Astronomical Almanac, including both positional
and physical data interpolated to any date and time within the appropriate year.
Versions are available for microcomputers running under MS-DOS and Macintosh
Systems.
Publications of the United States Naval Observatory, Second Series, are issued
irregularly and contain observational results or data related to obtaining observa-
tions.
14 / RELATED PUBLICATIONS
14.13 Other Publications of the Royal Greenwich Observatory
669
The Star Almanac for Land Surveyors contains tabulations of R, declination and
E for the Sun for every 6 hours, and right ascension to 0?1 and declination of 'l" of
all stars brighter than magnitude 4.0 for each month. In addition, the ephemerides
of R, declination, and E for the Sun are represented by polynomial series for each
month. This volume is available from Her Majesty's Stationery Office and from
Bernan Associates, 9730 E. George Palmer Highway, Lanham, MD, 20706.
Compact Data for Navigation and Astronomy for 1986 to 1990 contains data
which are mainly in the form of polynomial coefficients, for use by navigators and
astronomers to calculate the positions of the Sun, Moon, navigational planets, and
bright stars using a small programmable calculator or personal computer.
Interpolation and Allied Tables contains tables, formulas, and explanatory notes
on the techniques for numerical interpolation, differentiation, and integration. In
particular, it contains extensive tables of Bessel and Everett interpolation coeffi-
cients. This booklet is available from Her Majesty's Stationery Office and Bernan
Associates, 9730 E. George Palmer Highway, Lanham, MD, 20706. The companion
booklet Subtabulation contains tables of Lagrange interpolation coefficients at in-
tervals of 1/20 and 1/24, as well as details for two other techniques of systematic
interpolation.
Royal Observatory Bulletins (21-181) and Royal Greenwich Observatory Bul-
letins (1-20 and from 182) are issued regularly and contain details of current astro-
nomical research. These publications may be obtained, subject to availability, from
the Royal Greenwich Observatory, Madingly Road, Cambridge CB3 0EZ.
Selected NAO Technical Notes of the Royal Greenwich Observatory are issued
irregularly to disseminate technical information concerning astronomical phenom-
ena, ephemerides, and navigation.
Greenwich Observations is a complete list of the appendices and special in-
vestigations included in the annual volumes of observations made at the Royal
Observatory, Greenwich. A list of the separate publications of the Observatory are
given in the volume for 1946.
Royal Observatory Annals This series of publications includes: Number 1, "Nu-
tation 1900-1959," 1961; values based on E.W. Woolard's series.
Annals of Cape Observatory This series includes papers and observational data
that are of relevance to the ephemerides.
14.14 Publications of Other Countries
Apparent Places of Fundamental Stars is prepared annually by the Astronomis-
ches Rechen-Institut in Heidelberg and contains mean and apparent coordinates of
1535 stars of theFifth Fundamental Catalogue (FK5). This volume is available from
Verlag G. Braun, Karl-Friedrich-Strasse, 14-18, Karlsruhe, Germany.
qjq EXPLANATORY SUPPLEMENT
Ephemerides of Minor Planets is prepared annually by the Institute of The-
oretical Astronomy, and published by the Academy of Sciences of the U.S.S.R.
Included in this volume are elements, opposition dates, and opposition ephemer-
ides of all numbered minor planets. This volume is available from the Institute for
Theoretical Astronomy, 10 Kutuzov Quay, 191187 St. Petersburg.
14.2 ASTRONOMICAL PAPERS PREPARED FOR THE USE OF THE
AMERICAN EPHEMERIS AND NAUTICAL ALMANAC
Astronomical Papers of The American Ephemeris are available from the U.S. Naval
Observatory, Washington, DC, 20392. Volumes I through X are out of print and no
longer available.
Introduced in 1882, the series is published irregularly and contains reports of
research in celestial mechanics with particular relevance to ephemerides. A full list
of the papers published to date follows:
Volume I
I. Newcomb, Simon "On the Recurrence of Solar Eclipses with Tables of Eclipses
from B.C. 700 to A.D. 2300" 1879.
II. Newcomb, Simon, assisted by John Meier "A Transformation of Hansen's
Lunar Theory Compared with the Theory of Delaunay" 1880.
III. Michelson, Albert A. "Experimental Determination of the Velocity of Light
Made at the U.S. Naval Academy, Annapolis" 1880.
IV. Newcomb, Simon "Catalogue of 1098 Standard Clock and Zodiacal Stars"
1882.
V. Hill, George W. "On Gauss's Method of Computing Secular Perturbations,
with an Application to the Action of Venus on Mercury" 1881.
VI. Newcomb, Simon "Discussion and Results of Observations on Transits of
Mercury, from 1677 to 1881" 1882.
Volume II
I. Newcomb, Simon, assisted by John Meier "Formulae and Tables for Express-
ing Corrections to the Geocentric Place of a Planet in Terms of Symbolic
Corrections to the Elements of the Orbits of the Earth and Planet" 1883.
II. Safford, Truman Henry "Investigation of Corrections to the Greenwich Plan-
etary Observations from 1762 to 1830" 1883.
14 / RELATED PUBLICATIONS
671
III. Newcomb, Simon "Measures of the Velocity of Light Made under Direction
of the Secretary of the Navy During the Years 1880-1882" 1885.
IV. Michelson. Albert A. "Supplementary Measures of the Velocities of White
and Colored Light in Air, Water, and Carbon Disulphide, Made with the
Aid of the Bache Fund of the National Academy of Sciences" 1885.
V. Newcomb, Simon "Discussion of Observations of the Transits of Venus in
1761 and 1769" 1890.
VI. Newcomb, Simon "Discussion of the North Polar Distances Observed with
the Greenwich and Washington Transit Circles with Determinations of the
Constant of Nutation" 1891.
Volume III
I. Newcomb, Simon "Development of the Perturbative Function and its Deriva-
tives, in Sines and Cosines of Multiples of the Eccentric Anomalies, and in
Powers of the Eccentricities and Inclinations" 1884.
II. Hill, G.W. "Determination of the Inequalities of the Moon's Motion Which
are Produced by the Figure of the Earth: A Supplement to Delaunav's
Lunar Theory" 1884.
III. Newcomb, Simon "On the Motion of Hyperion. A New Case in Celestial
Mechanics" 1884.
IV. Hill, G.W. "On Certain Lunar Inequalities Due to the Action of Jupiter
and Discovered by Mr. E. Neison" 1885.
V. Newcomb, Simon "Periodic Perturbations of the Longitudes and Radii Vec-
tors of the Four Inner Planets of the First Order as to the Masses" 1891.
Volume IV
Hill, G.W. "A New Theory of Jupiter and Saturn" 1890.
Volume V
I. Newcomb, Simon "A Development of the Perturbative Function in Cosines of
Multiples of the Mean Anomalies and of Angles Between the Perihelia and
Common Node and in Powers of the Eccentricities and Mutual Inclination"
1895.
II. Newcomb, Simon "Inequalities of Long Period, and of the Second Order as
to the Masses, in the Mean Longitudes of the Four Inner Planets" 1895.
III. Newcomb, Simon "Theory of the Inequalities in the Motion of the Moon
Produced by the Action of the Planets" 1895.
IV. Newcomb, Simon "Secular Variations of the Orbits of the Four Inner Plan-
ets" 1895.
V. Newcomb, Simon "On the Mass of Jupiter and the Orbit of Polyhymnia"
1895.
672
Volume VI
EXPLANATORY SUPPLEMENT
Tables of the four inner planets.
I. Newcomb, Simon "Tables of the Motion of the Earth on its Axis and Around
the Sun" 1895.
II. Newcomb, Simon "Tables of the Heliocentric Motion of Mercury" 1895.
III. Newcomb, Simon "Tables of the Heliocentric Motion of Venus" 1895.
IV. Newcomb, Simon "Tables of the Heliocentric Motion of Mars" 1898.
Volume VII
I. Hill, George William "Tables of Jupiter, Constructed in Accordance with the
Methods of Hansen, and Prepared for Use in the Office of the American
Ephemeris and Nautical Almanac" 1895.
II. Hill, George William "Tables of Saturn, Constructed in Accordance with the
Methods of Hansen, and Prepared for Use in the Office of the American
Ephemeris and Nautical Almanac" 1898.
III. Newcomb, Simon "Tables of the Heliocentric Motion of Uranus" 1898.
IV. Newcomb, Simon "Tables of the Heliocentric Motion of Neptune" 1898.
Volume VIII
I. Newcomb, Simon "A New Determination of the Precessional Constant with
the Resulting Precessional Motions" 1897.
II. Newcomb, Simon "Catalogue of Fundamental Stars for the Epochs 1875 and
1900 Reduced to an Absolute System" 1899.
III. Hedrick, Henry B. "Catalogue of Zodiacal Stars for the Epochs 1900 and
1920 Reduced to an Absolute System" 1905.
Volume IX
I. Newcomb, Simon "Researches on the Motion of the Moon. Part II. The Mean
Motion of the Moon and Other Astronomical Elements Derived from Ob-
servations of Eclipses and Occupations Extending from the Period of the
Babylonians until A.D. 1908" 1912.
II. Ross, Frank E. "New Elements of Mars and Tables for Correcting the He-
liocentric Positions Derived from Astronomical Papers, Vol. VI, Part IV"
1917.
III. Eichelberger, W.S. and Newton, Arthur "The Orbit of Neptune's Satellite
and the Pole of Neptune's Equator" 1926.
Volume X
I. Eichelberger, W.S. "Positions and Proper Motions of 1504 Standard Stars
for the Equinox 1925.0" 1925.
14 / RELATED PUBLICATIONS
673
II. Robertson, James "Catalog of 3539 Zodiacal Stars for the Equinox 1950.0"
1940.
Volume XI
I. Clemence, G.M. "The Motion of Mercury 1765-1937" 1943
II. Clemence, G.M. "First-order Theory of Mars" 1949
III. Morgan, H.R. "Definitive Positions and Proper Motions of Primary Refer
ence Stars for Pluto" 1950. V r"
IV. Herget Paul, Clemence, G.M., and Hertz, Hans G. "Rectangular Coordi-
nates oi Ceres, Pallas, Juno, Vesta 1920-1960" 1950.
Volume XII
Eckert, W.J., Brouwer, Dirk, and Clemence, G.M. "Coordinates of the Five
Outer Planets 1653-2060" 1951.
Volume Xffl
I. van Woerkom, A.J.J. "The Motion of Jupiter's Fifth Satellite 1892-1949"
1950.
II. Brouwer Dirk and van Woerkom, A.J.J. "The Secular Variations of the
Orbital Elements of the Principal Planets" 1950
III. Morgan, H.R. "Catalog of 5268 Standard Stars, 1950.0. Based on the Nor-
mal System N30" 1952.
IV. Clemence, G.M. "Coordinates of the Center of Mass of the Sun and the
Five Outer Planets, 1800-2060" 1953.
V. Clemence, G.M. "Perturbations of the Five Outer Planets by the Four Inner
Ones" 1954.
Volume XIV
Herget, Paul "Solar Coordinates 1800-2000" 1953.
Volume XV
I. Woolard, Edgar W. "Theory of the Rotation of the Earth Around Its Center
of Mass" 1953.
II. Hertz, Hans G. "The Mass of Saturn and the Motion of Jupiter 1884-1948"
1953.
III. Herget, Paul "Coordinates of Venus 1800-2000" 1955.
Volume XVI
I. Duncombe, Raynor L. "Motion of Venus 1750-1949" 1958.
II. Clemence, G.M. "Theory of Mars-Completion" 1961.
III. Herget. Paul "Rectangular Coordinates of Ceres, Pallas. Juno Vesta 1960
1980" 1962.
674 EXPLANATORY SUPPLEMENT
Volume XVII
Watts, C.B. "The Marginal Zone of the Moon" 1963.
Volume XVIII
Jarnagin, Jr., Milton P. "Expansions in Elliptic Motion" 1965.
Volume XIX
I. Franz. Otto G. and Mintz, Betty F. "Tables of X and Y- Elliptic Rectangular
Coordinates" 1964.
II. Eckert, W.J. and Smith, Jr., Harry F. "The Solution of the Main Problem
of the Lunar Theory by the Method of Airy" 1966.
Volume XX
L Jackson, Edward S. "Determination of the Equinox and Equator from Merid-
ian Observation of the Minor Planets" 1968.
II. Duncombe, Raynor L. "Heliocentric Coordinates of Ceres, Pallas, Juno,
Vesta 1928-2000" 1969.
III. O'Handley, Douglas A. "Determination of the Mass of Jupiter from the
Motion of 65 Cybele" 1969.
Volume XXI
I. Janiczek, Paul M. "The Orbit of Polyhymnia and the Mass of Jupiter" 1971.
II. Fiala, Alan D. "Determination of the Mass of Jupiter from a Study of the
Motion of 57 Mnemosyne" 1972.
III. Branham, Jr., Richard L. "The Orbits of Five Minor Planets and Correc-
tions to the FK4 Equator and Equinox" 1979.
Volume XXII
I. Cohen, C.J., Hubbard, E.C., and Oesterwinter, Claus "Elements of the Outer
Planets for One Million Years" 1973.
II. Jackson, Edward S. "A Discussion of the Observations of Neptune 1846
1970" 1974.
III. Pierce, David A. "Star Catalog Corrections Determined from Observations
of Selected Minor Planets" 1978.
IV. Laubscher, Roy Edward "The Motion of Mars 1751-1969" 1981.
Volume XXIII
I. Gutzwiller, Martin C. and Schmidt, Dieter S. "The Motion of the Moon as
Computed by the Method of Hill, Brown, and Eckert" 1986.
14 / RELATED PUBLICATIONS
14.3 UNITED STATES NAVAL OBSERVATORY CIRCULARS
675
United States Naval Observatory Circulars are issued irregularly to disseminate
astronomical data concerning ephemerides or astronomical phenomena.
Circulars are available from the U.S. Naval Observatory, Washington, DC 20392.
Copies of the first 100 circulars are no longer available. Circulars from 101 onward
are available with the exceptions of 103, 104, 109, 115, 118, and 138. The content
of the first one hundred circulars can be summarized as follows:
Subject
Circular tt
Sunspots. positions, areas, counts
3, 4, 6, 8-13, 15, 17, 19, 21-26, 28-39. 41-48, 50-52, 54-58 60-77
79-84, 86, 87
Eclipses, 1952-1964
1,2, 16, 27, 40, 53, 59, 78, 85, 88, 89
Minor Planet Ephemerides
5, 7, 18, 20
Naval Observatory Time Service
14, 49
Ephemeris of Mars
90, 95. 98
Coordinates of the Moon
91
Radio Longitude of the Meridian of Jupiter
92,94
GC and DM Numbers
93
Distance and Velocity of Venus
96
Polaris and Equation of Time
97
Machine Readable Data
99
Index
100
United States Naval Observatory Circulars, #101 onward
No. Title
101 Solar Eclipses, 1971-1975
1 02 Total Solar Eclipse of 30 May 1 965
103 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
104 Durchmusterung and Henry Draper Numbers of Albany General Catalog Stars
105 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
1 06 Rectangular Coordinates of Mercury 1 800-2000
1 07 Sunlight, Moonlight , and Twilight for Antarctica 1 966-1 968
108 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
1 09 Annular Solar Eclipse of 20 May 1 966
1 1 0 Total Solar Eclipse of 1 2 November 1 966
1 1 1 Astronomical Data in Machine Readable Form
1 1 2 Phases of the Moon 1 800-1 959
113 Solar Eclipses, 1976-1980
114 Astronomical Data in Machine Readable Form
115 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
1 1 6 Total Solar Eclipse of 22 September 1 968
1 1 7 Ephemeris of trie Radio Longitude of the Central Meridian of Jupiter System III (1957.0)
118 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
1 1 9 Phases of the Moon 1 960-2003
120 Sunlight, Moonlight, and Twilight for Antarctica 1969-1971
121 A Linotron System Manual for the Photo-Composition of Astronomical and Mathematical Tables
122 Annular Solar Eclipse of 18 March 1969
123 Annular Solar Eclipse of 1 1 September 1969
676
United States Naval Observatory Circulars, continued
EXPLANATORY SUPPLEMENT
No. Title
124 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
1 25 Total Solar Eclipse of 7 March 1 970
126 Annular Solar Eclipse of 31 August-1 September 1970
127 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
128 Astronomical Data in Machine Readable Form
129 Annular Solar Eclipse of 16 January 1972
1 30 Perigee and Apogee of the Moon 1 959-1 999
131 Total Solar Eclipse of 1 0 July 1 972
132 Sunlight, Moonlight, and Twilighl lor Antarctica 1972-1974
1 33 Sunspot Areas 1 907-1 970
134 Normalized Observations of Venus 1901-1949
1 35 Total Solar Eclipse of 30 June 1 973
1 36 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
137 Ephemeris of the Radio Longitude of the Central Meridian of Jupiter System III (1957.0)
1 38 Geocentric Solar Data
139 Astrometric Ephemeris of Pluto 1970-1990
1 40 Rectangular Coordinates of the Moon 1 971 -1 980
141 Lunar Limb Profiles for Solar Eclipses
1 42 Solar Eclipses, 1 981-1 990
143 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
144 Total Solar Eclipse Of 20 June 1974
1 45 Normalized Observations of Mercury 1901-1937
146 Astronomical Data in Machine Readable Form
147 Sunlight, Moonlight, and Twilight for Antarctica 1975-1977
148 Physical Ephemeris of Mars
149 Fortran Automatic Typesetting System
150 Index to United States Naval Observatory Circulars Nos. 101-150
1 5 1 Apparent Ephemeris of Mars 1 960-1 980
1 52 Total Solar Eclipse of 23 October 1 976
153 Coordinates of U.S. Naval Observatory Installations
154 The Determination of Universal Time at the U.S. Naval Observatory
1 55 Almanac for Computers , 1 977
1 56 Solar Eclipses of 1 977
157 Total Solar Eclipse ot 26 February 1979
1 58 Total Solar Eclipse of 1 6 February 1 980
159 Observations of the Sun, Moon, and Planets/Six-lnch Transit Circle Results
1 60 Total Solar Eclipse of 31 July 1 981
161 Ephemeris of the System III (1965) Longitude of the Central Meridian of Jupiter
162 Geocentric Ephemeris of Pluto 1975-1985
163 The IAU Resolutions on Astronomical Constants, Time Scales, and the Fundamental Reference Frame
164 Astronomical Data in Machine Readable Form
1 65 Total Solar Eclipse of 1 1 June 1 983
166 Annular Solar Eclipse of 30 May 1 984
167 Project Merit Standard
1 68 Total Solar Eclipse of 22-23 November 1 984
1 69 Phases of the Moon 2000-2049
170 Solar Eclipses, 1991-2000
171 Computer Programs for Sun and Moon Illuminance with Contingent Tables and Oiagrams
1 72 Total Solar Eclipse of 1 7- 1 8 March 1 988
1 73 Total Solar Eclipse of 22 July 1 990
1 74 Total Solar Eclipse of 1 1 July 1 991
1 75 Annular Solar Eclipse of 1 5-1 6 January 1 991
1 76 Central Solar Eclipses ot 1 992
177 On High Frequency Spectral Analysis of Very Noisy Stochastic Processes and Its Application to ERP
Series Determined by VLBI and SLR
14 / RELATED PUBLICATIONS 677
1 4.4 PUBLICATIONS OF THE UNITED STATES NAVAL OBSERVATORY,
SECOND SERIES
The Publications of the United States Naval Observatory, Second Series are issued
irregularly to disseminate observational results. These publications, from Volume
XVII onward, are available from the U.S. Naval Observatory. Washington, DC
20392.
Volume I
Harkness, W. and Skinner, A.N. "Transit Circle Observations of the Sun,
Moon, Planets, and Miscellaneous Stars, 1894-1899" 1900.
Volume II
Skinner, A.N. "Zone Observations with the Nine-Inch Transit Circle, 1894-
1901" 1902.
Volume III
I. See, T.J.J. "Observations of Eros with the Twenty-Six Inch Equatorial, 1900-
1901" 1903.
II. King, T.I. "Observations of Eros and Reference Stars with the Nine-Inch
Transit Circle, 1900-1901" 1903.
III. Eichelberger, W.S. "Observations of 495 Zodiacal Stars with the Nine-Inch
Transit Circle, 1900" 1903.
IV. Updegraff, M. "Observations with the Six-Inch Transit Circle, 1900-1901"
1903.
V. Ingersoll, R.R., Bowman, C.G., and Taylor, H. "Observations with the Prime
Vertical Transit Instrument, 1882-1884" 1903.
Volume IV
L "Transit Circle Observations of the Sun, Moon, Planets, and Miscellaneous
Stars, 1900-1903" 1906.
II. Eastman, J.R. "Transit Circle Observations of the Sun, Moon, Planets, and
Comets, 1866-1891" 1906.
III. Updegraff, M. "Observations with the Six-Inch Transit Circle, 1901-1902"
1906.
IV. Appendices
I. "Total Solar Eclipses of May 28, 1900, and May 17, 1901" 1906.
II. Eichelberger, W.S. "Reduction Tables for Transit Circle Observations"
1906.
III. Frederick, C.W. "Reduction Tables for Equatorial Observations" 1906.
IV. Hayden, Edward Everett "The Present Status of the Use of Standard
Time" 1906.
678
Volume V
EXPLANATORY SUPPLEMENT
"Meteorological Observations and Results, 1893-1902" 1903.
Volume VI
"Equatorial Observations, 1893-1907" 1911.
Appendices
I. "Miscellaneous Astronomical Papers by Members of the Naval Observa-
tory Staff" 1911.
II. "Miscellaneous Reports on the Transit of Mercury of November 10,
1894" 1911.
III. Horigan, Wm.D. "List of Publications Issued by the United States
Naval Observatory, 1845-1908" 1911.
Volume VII
Eichelberger, W.S. and Littell, F.B. "Catalogue of 23521 Stars between 13°35'
and 45°25' South Declination for the Equinox 1850 from Zone Observations
Made at the United States Naval Observatory, 1846-1852" 1911.
Volume VIII
Littell, F.B., Hill, G.A., and Evans, H.B. "Vertical Circle Observations Made
with the Five-Inch Alt-Azimuth Instrument, 1898-1907" 1914.
Volume IX
I. Eichelberger, W.S. and Morgan, H.R. "Results of Observations with the Nine-
Inch Transit Circle, 1903-1911" 1920.
II. Eichelberger, W.S. "Observations Made with the Nine-Inch Transit Circle,
1903-1908" 1915.
III. Littell, F.B. and Eichelberger, W.S. "Observations Made with the Nine-
Inch Transit Circle, 1908-1911" 1918.
IV. Littell, F.B. and Morgan, H.R. "Observations Made with the Nine-Inch
Transit Circle, 1912-1913" 1918.
Appendix:
Littell, F.B. and Hill, G.A. "Determination of the Difference of Longitude
Between Washington and Paris, 1913-1914" 1918.
Volume X
I. Hill, G.A. "Observations Made with the Prime Vertical Transit Instrument,
1893-1912" 1926.
II. "Total Solar Eclipses of August 30, 1905, and June 8, 1918, with Aviators'
Notes on the Total Solar Eclipse of September 10, 1923" 1926.
14 / RELATED PUBLICATIONS
Volume XI
679
Hammond, J,C, and Watts, C.B. "Results of Observations with the Six-Inch
Transit Circle, 1909-1918" 1927.
Volume XII
I. "Equatorial Observations, 1908-1926" 1929.
II. "Photographic Equatorial Observations, 1912-1924 and Photoheliographic
Observations, 1917-1927" 1929.
Appendix:
Littell, F.B., Hammond, J.C., Watts, C.B., and Sollenberger, P. "World
Longitude Operation of 1926 — Results of Observations at San Diego
and Washington" 1929.
Volume Xm
Morgan, H.R. "Results of Observations with the Nine-Inch Transit Circle,
1913-1926. Observations of the Sun, Moon, and Planets. Catalogue of 9,989
Standard and Intermediary Stars. Miscellaneous Stars" 1933.
Appendices
I. Vening Meinesz, F.A. and Wright, F.E. "The Gravity Measuring Cruise
of the U.S. Submarine S-21 with an Appendix on Computational Pro-
cedure" 1930.
II. Lamson, Eleanor A. "Declinations of Stars Derived from Observations
with the Prime Vertical Transit Instrument, 1921-1925" 1932.
III. "Total Solar Eclipses of January 24, 1925, January 14, 1926 and May
9, 1929" 1932.
IV. Morgan, H.R. "Washington Observations of the Moon, 1894-1922"
1932.
Volume XIV
I. Hammond, J.C. "Catalog of 3,520 Zodiacal Stars Based on Observations with
the Six-Inch Transit Circle, 1928-1930 Reduced Without Proper Motion to
the Equinox 1925.0" 1938.
II. Morgan, H.R. and Lyons, U.S. "Results of Observations on the Nine-Inch
Transit Circle, 1932-1934. Positions and Proper Motions of 1,117 Reference
Stars in Declination -10° to ^20°. Miscellaneous Stars" 1938.
III. Littell, F.B., Morgan, H.R., and Raynsford, G.M. "Vertical Circle Observa-
tions Made with the Five-Inch Alt-Azimuth Instrument, 1916-1933. Cat-
alog of Declinations of Standard Stars. Declinations of the Sun, Mercury,
and Venus" 1938.
IV. Morgan, H.R. "Proper Motions of 2,916 Intermediary Stars Mostly in Dec-
lination -5° to -30°" 1938.
680
EXPLANATORY SUPPLEMENT
V. Watts, C.B., Sollenberger, P., and Willis, J.E. "World Longitude Operation
of 1933 at San Diego and Washington" 1938.
Volume XV
I. Wylie, L.R. "A Comparison of Newcomb's Tables of Neptune with Observa-
tion, 1795-1938" 1942.
II. Clemence, G.M. and Whittaker, G.C. "Observations of the Transit of Mer-
cury November 11-12, 1940" 1942.
III. Wylie, L.R. "An Investigation of Newcomb's Theory of Uranus" 1947.
IV. Burton, H.E., Lyons, U.S., Raynsford, G.M., Wylie, L.R., Browne, W.M.,
and Smith, J.L. "Observations of Double Stars Made with the 26-Inch
Equatorial of the U.S. Naval Observatory, 1928-44" 1947.
V. Morgan H.R. and Scott, F.P. "Results of Observations Made with the Nine-
Inch Transit Circle, 1935-1945. Observations of the Sun and Planets. Cat-
alog of 5,446 Stars. Corrections to GC and FK3" 1948.
Volume XVI
I. Watts, OB. and Adams, A.N. "Results of Observations Made with the Six-
Inch Transit Circle, 1925-1941. Observations of the Sun, Moon, and Plan-
ets. Catalog of 2,383 Stars for 1925.0. Catalog of 1,536 Stars for 1950.0.
Corrections to GC and FK3" 1949.
II. Watts, OB. "Description of the Six-Inch Transit Circle. Instrumental De-
velopments, 1932-1948" 1950.
III. Watts, C.B., Scott, F.P., and Adams, A.N. "Results of Observations Made
with the Six-Inch Transit Circle, 1941-1949. Observations of the Sun, Moon,
and Planets. Catalog of 5,216 Stars for 1950.0. Corrections to GC and FK3"
1952.
Volume XVII
I. Hall, John S. and Mikesell, A.H. "Polarization of Light in the Galaxy as
Determined from Observations of 551 Early- Type Stars" 1950.
II. Lyons, U.S. and Wylie, L.R. "Observations of Double Stars Made with the
26-Inch Equatorial of the U.S. Naval Observatory, 1945-1948" 1953.
III. Hall, Jr., A., Burton, H.E., Lyons, U.S., Wylie, L.R., Raynsford, G.M.,
Browne, W.M., and Smith, J.L. "Observations of Satellites Made with the
26-Inch Equatorial of the U.S. Naval Observatory, 1928-1947" 1953.
IV. Mikesell, A.H. "The Scintillation of Starlight" 1955.
V. Markowitz, W. "Observations of Double Stars, 1949-52, and a Study of the
Optics of the 26-Inch Refractor" 1956.
VI. Hall, J.S. "Polarization of Starlight in the Galaxy" 1958.
VII. Hoag, A.A., Johnson, H.L., Iriarte, B., Mitchell, R.I., Hallam, K.L., and
Sharpless, S. "Photometry of Stars in Galactic Cluster Fields" 1961.
14 / RELATED PUBLICATIONS
Volume XVI 1 1
681
I. Franz, O.G., Gossner, J.L., Josties, F.J., Lindenblad, I.W., Mikesell, A.H.
Mintz, B.F., and Riddle, R.K. "Photographic Measures of Double Stars"
1963.
IIA. Mikesell, A.H. "Mechanical Improvement of the 26-Inch Refractor" 1968.
IIB. Riddle, R.K, "A Study of the Optical Properties of the 26-Inch Refractor"
1968.
III. Worley, C.E. "A Catalog of Visual Binary Orbits" 1963.
IV. Walker, Jr., R.L. "Micrometer Measures of 256 Double Stars" 1966.
V. Strand, K. Aa. "Photovisual Magnitude Differences of Double Stars" 1969.
VI. Worley, C.E. "Micrometer Measures of 1164 Double Stars" 1967.
VII. Kallarakal, V.V, Lindenblad, I.W., Josties, F.J, Riddle, R.K, Miranian,
M, Mintz, B.F, and Klugh, A.P. "Photographic Measures of Double Stars"'
1969.
Volume XIX
L Adams, A.N, Bestul, S.M, and Scott, D.K. "Results of Observations Made
with the Six-Inch Transit Circle, 1949-1956. Observations of the Sun, Moon,
and Planets. Catalog of 5,965 Stars for 1950.0. Corrections to FK4 Gc'
and N30" 1964.
II. Adams, A.N. and Scott, D.K. "Results of Observations Made with the Six-
Inch Transit Circle, 1956-1962. Observations of the Sun, Moon, and Plan-
ets. Catalog of 2,554 Stars for 1950.0. Corrections to FK4, GC and N30"
1968.
III. Adams, A.N, Klock, B.L, and Scott, D.K. "Washington Meridian Obser-
vations of the Moon. Six-Inch Transit Circle Results, 1925-1968" 1969.
Volume XX
I. Strand, K. Aa. "The 61-Inch Astrometric Reflector System" 1971.
II. Hoag, A. A, Priser, J.B, Riddle, R.K, and Christy, J.W. "Installation,
Tests, and Initial Performance of the 61-Inch Astrometric Reflector" 1967.
IIIA. Riddle, R.K. "First Catalog of Trigonometric Parallaxes of Faint Stars—
Astrometric Results" 1970.
IIIB. Priser, J.B. "First Catalog of Trigonometric Parallaxes of Faint Stars-
Photometric Results" 1970.
IIIC. Strand, K. Aa. and Riddle, R.K. "First Catalog of Trigonometric Paral-
laxes of Faint Stars — Discussion" 1970.
IV. Abies, H.D. "Optical Study of Nearby Galaxies" 1971.
V. Kron, G.E, Guetter, H.H, and Riepe, B.Y. "A Catalog of Colorimetric
Measures of Stars on the Six-Color System of Stebbins and Whitford"
1972.
VI. Routly, P.M. "Second Catalog of Trigonometric Parallaxes of Faint Stars"
1972.
EXPLANATORY SUPPLEMENT
VII. Priser, J.B. "UBV Sequences in Selected Areas" 1974.
Volume XXI
Blanco, V.M., Demers, S., Douglass, G.G., and Fitzgerald, MP. "Photoelectric
Catalogue. Magnitudes and Colors of Stars in the U, B, V and Uc, B, V
Systems" 1968.
Volume XXII
i. Walker, Jr., R.L. "Micrometer Measures of 463 Double Stars" 1969.
II. Worley, C.E. "Micrometer Measures of 1,343 Double Stars " 1971.
III. Harrington, R.S, and Mintz, B.F. "Positions of Bright Minor Planets" 1972.
IV. Worley, C.E. "Micrometer Measures of 1,056 Double Stars" 1972.
V. Walker, Jr., R.L. "Micrometer Measures of 618 Double Stars" 1972.
VI. Josties, F.J., Dahn, C.C., Kallarakal, V. V., Miranian, M., Douglass, G.G.,
Christy, J.W., Behall, A.L., and Harrington, R.S. "Photographic Measures
of Double Stars" 1974.
Volume XXm
I. Rhynsburger, RW. and Gauss, F.S. "Catalog of Proper Motions for the 5,965
Stars of the Six-Inch Transit Circle Program, 1949-1956" 1975.
H. Rhynsburger, R.W., Gauss, F.S., and Crull, Jr., H.E. "Motions and Phys-
ical Properties of 1,086 Cepheids and Early- Type Stars Observed in the
Washington Six-Inch Transit Circle Program, 1949-1956" 1980.
III. Hughes, J. A. and Scott, D.K. "Results of Observations Made with the
Six-Inch Transit Circle, 1963-1971. Observations of the Sun, Moon, and
Planets. Catalog of 14,916 Stars for 1950.0. Comparison with FK4, GC,
and N30" 1982.
Volume XXIV
I. Harrington, R.S., Dahn, C.C., Behall, A.L., Priser, J.B., Christy, J.W., Riepe,
B.Y., Abies, H.D., Guetter, H.H., Hewitt, A.V., and Walker, R.L. "Third
Catalog of Trigonometric Parallaxes of Faint Stars" 1975.
II. Behall, A.L. "Micrometer Measures of 267 Double Stars" 1976.
III. Dahn, C.C., Harrington, R.S., Riepe, B.Y., Christy, J.W., Guetter, H.H.,
Behall, A.L., Walker, R.L., Hewitt, A.V., and Abies, H.D. "Fourth Catalog
of Trigonometric Parallaxes of Faint Stars" 1976.
IV. Harrington, R.S., Dahn, C.C., Miranian, M., Riepe, B.Y., Christy, J.W.,
Guetter, H.H., Abies, H.D., Hewitt, A.V., Vrba, F.J., and Walker, R.L.
"Fifth Catalog of Trigonometric Parallaxes of Faint Stars" 1978.
V. Josties, F.J., Kallarakal, V.V., Douglass, G.G., and Christy, JW. "Photo-
graphic Measures of Double Stars" 1978.
14 / RELATED PUBLICATIONS
683
VI. Worley, C.E. "Micrometer Measures of 1,980 Double Stars" 1978.
VII. Worley, C.E. and Heintz, W.D. "Fourth Catalog of Orbits of Visual Binary
Stars" 1983.
Volume XXV
I. Rydgren, A.E,, Schmelz, J.T., Zak, D.S., and Vrba, F.J. "Broad Band Spec-
tra] Energy Distributions of T Tauri Stars in the Taurus-Auriga Region"
1984.
II. Walker, Jr., R.L. "Micrometer Measures of 711 Double Stars" 1985.
III. Worley, C.E. "Micrometer Measures of 2589 Double Stars" 1989.
Volume XXVI
I. Scott, D.K. "Results of Observations Made with the Six-Inch Transit Circle,
1956-1962. Catalog of 10,010 AGK3R Stars for 1950.0. Comparison with
GC and N30" 1987.
II. Hughes, J. A., Smith, OA., and Branham, R.L. "Results of Observations
Made with the Seven-Inch Transit Circle, 1967-1973" 1992.
684
EXPLANATORY SUPPLEMENT
14.5 SELECTED NAO TECHNICAL NOTES OF THE ROYAL GREENWICH
OBSERVATORY
The NAO Technical Notes are published irregularly by the Royal Greenwich Ob-
servatory to disseminate technical information of specific interest.
No.
Author
Title
Year
6 Hobden, D.E.
11 Sadler, D.H.
12
18
26
29
31
36
37
47
48
53
54
55
56
57
58
62
63
64
65
Scott, W.A.
Sadler, D.H.
Taylor, G.E.
Sinclair, A.T.
Taylor, G.E.
Wilkins, G.A.
Sinclair, A.T.
Sinclair, A.T.
39 Watson, F.G.
40 Morrison, L.V.
46 Yallop, B.D.
Emerson, B.
Emerson, B.
Sinclair, A.T.
Yallop, B.D.
Yallop, B.D.
Emerson, B.
Emerson, B.
Yallop, B.D.
Yallop, B.D.
59 Sinclair, A.T.
Yallop, B.D.
Hohenkerk, C.Y.
Hohenkerk, C.Y,
Sinclair, A.T.
Yallop, B.D.
Hohenkerk, C.Y.
Computations for the Decca Hyperbolic Lattice 1966
System
Interpolation in H.0. 299, Sight Reduction Tables for 1966
Marine Navigation
Corrections, for Irradiation, to the Observed Altitude 1 967
of the Sun
Predictions of Grazing Occultations 1970
The Representation of Planetary Ephemerides for 1971
use in the Automatic Reduction of Navigational
Sights
The Visual Observation of Occultations 1974
The Future Publication of Astronomical 1974
Ephemerides
The Orbits of Tethys, Dione, Rhea, and Titan 1974
The Tabulation of Monthly Sets of Polynomial Coef- 1 975
ficients for Astronomical Data in the Star Almanac
for Land Surveyors
The Zenithal Blind Spot of a Large Altazimuth 1976
Telescope
The Movements of the Sun, Moon, and Stars in the 1976
Sky
Formulae for Computing Astronomical Data with 1978
Hand-Held Calculators
Approximate Solar Coordinates 1 978
Approximate Lunar Coordinates 1979
The Computation of Physical Ephemerides of Plan- 1 980
ets and Satellites
Geocentric and Heliocentric Phenomena 1981
The Phase Correction for Venus 1 981
Approximate Coordinates of Jupiter and Saturn 1981
Ground Illumination 1986
Formulae for Determining Carrington's Elements 1982
and Differential Solar Rotation
The Effect of Atmospheric Refraction on Laser 1982
Ranging Data
Coefficients for Calculating the GHA and DEC of 1985
Stars
The Computation of Angular Atmospheric Refrac- 1 985
tion at Large Zenith Angles
Algorithms for Calculating the Dates of Easter 1 986
Determination of Polynomial Coefficients from B- 1986
Spline Coefficients
14 / RELATED PUBLICATIONS
14.6 LISTS OF APPENDICES AND SUPPLEMENTS
685
14.61 The British Nautical Almanac
Many issues, especially the earlier years, of the almanac contain appendices on
various astronomical and navigational subjects. A list of these and of separate sup-
plements, together with sections of a similar nature in the prefaces and explanations
to certain issues, is given below. Unaltered, or nearly unaltered, reprints in later
issues are ignored.
The appendices to the almanacs for 1772 to 1778, 1787, 1788, and 1794 were
collected and published in 1813 with the title Selections from the Additions That
Have Been Occasionally Annexed to The Nautical Almanac from its Commencement
to the Year 1812, while those to the almanacs for 1835 to 1854 were similarly
published in 1851 with the title Appendices to Various Nautical Almanacs Between
the Years 1834 and 1854.
1766: Maskelyne, N., et al. "Tables Requisite to be Used with The Astro-
nomical and Nautical Ephemeris." Separate publication, 166 pages.
1769: Maskelyne, N. "Instructions Relative to the Observation of the En-
suing Transit of the Planet Venus over the Sun's Disk, on the 3rd of
June 1769." 9 pages.
Maskelyne, N. "Use of the Astronomical Quadrant in Taking Alti-
tudes." 38 pages.
1771: Douwes, C. and Campbell, J. "Tables for... Finding the Latitude of a
Ship at Sea." 77 pages.
Maskelyne, N. "Determination [by John Bradley] of the Position of
the Lizard." 6 pages.
Wargentin, P.W. "Tabulae Novae et Correctae Pro Supputandis Eclipsi-
bus Tertii Satellites Jovis..." 16 pages.
1772: Maskelyne, N. "A Correct and Easy Method of Clearing the [Lunar]
Distance..." 25 pages.
"Eclipses of Jupiter's Third Satellite, and Tables of the Hour Angle
of the Sun and Jupiter." 6 pages.
Lyons, I. and Dunthorne, R. "Problems in Navigation." 8 pages.
1773: "A Table of the Equations to Equal Altitudes." 24 pages.
"A Catalogue of the Places of 387Fixed Stars,. ..(for 1760.0)." 14 pages.
1774: Mason, C. "Longitudes and Latitudes of the Moon, Deduced from
Dr. Bradley's Observations, Made Between September 13th 1750 and
November 2nd 1760, and Compared with a Set of Manuscript. Tables."
36 pages.
Maskelyne, N. "Elements of the Lunar Tables." 11 pages.
Maskelyne, N. "Remarks on the Hadley's Quadrant." 14 pages.
686 EXPLANATORY SUPPLEMENT
Lyons, I. "Astronomical Problems." 10 pages.
1778: Mason, C. "Right Ascensions and Zenith Distances of the Moon..."
40 pages.
Lyons, I. "Astronomical Problems." 11 pages.
1779: Wargentin, P.W. "Tabulae Novae et Correctae Pro Supputandis Eclipsi-
bus Secundi Satellitis Jovis..." 30 pages.
1781: Edwards, J. "Astronomical Problems." 27 pages.
Edwards, J. "Addition to the... Tables Annexed to the Nautical Al-
manac of 1771." 10 pages.
1787: Edwards. J. "Directions for Making... Reflecting Telescopes and...
Polishing... Them..." 48 pages.
Edwards, J. and Maskelyne, N. "An Account of the... Tremors of Re-
flecting Telescopes..." 12 pages.
1788: Blair, R. "Description of a.. .Method of Adjusting Hadley's Quad-
rant." 20 pages.
1791: Maskelyne, N. "Advertisement of the Expected... Comet of.. .1788.
and Relative to. ..Saturn's Ring in 1789 and 1790." 4 pages.
1794: Brinkley, J. "Tables to Improve. . .the Method of Finding the Latitude..."
15 pages.
1798: Brinkley. J. "Tables to Improve... Latitude..." (Second edition, re-
vised and corrected). 16 pages.
1809-1821: Each issue contains one or more catalogs of stars, some of longitude
and latitude, some of right ascension and declination; the number of
stars varies between 9 and about 50.
1812: Pond, .1. "On the Obliquity." 3 pages.
1818: Brinkley, J. "Two Practical Rules for Reducing the Observed Distance
of the Moon from the Sun or a... Star..." 18 pages.
1822: Brinkley, .1. "A Practical Method of Computing the Latitude." 16 pages.
1822 1833: "Tables of... Refraction,... of Second Differences, and... of Star Places."
16-40 pages.
1824-1833: "Elements of Occupations." 6-17 pages.
1826: "Rules for [Predicting] Occupations." 8 pages.
1827: Young, T. and Henderson, T. "Rules for [Reducing] Occupations."
4 pages.
1828-1833: Separately issued supplements for each of these years contained a
number of quantities that were transferred, between 1832 and 1834,
to the pages of the almanac proper,
1829: Lax, W. "An Easy Method of Finding the Latitude and Time at
Sea..." 23 pages.
1831: Lax, W. "An Easy Method of Correcting the Lunar Distance..."
6 pages.
14 / RELATED PUBLICATIONS Qqj
1832: Airy, G.B. "Corrections of the Longitudes and Right Ascensions of
the Sun..." 4 pages.
Jenkins. H. "Recalculated Elements of Delambre's Tables... of Jupiter's
Satellites." 12 pages.
1833: Schumacher, H.C. "Ephemeris of.. .Lunar Distances of Venus, Mars,
Jupiter, and Saturn." 44 pages.
"Geocentric Places of the Planets." 75 pages.
1834: "Report of the Committee of the Astronomical Society of London."
11 pages.
1835: Woolhouse, W.S.B. "New Tables for... Jupiter's Satellites..." 39 pages.
Woolhouse, W.S.B. "On the Computation of an Ephemeris of a Comet...'
9 pages.
"Comparison of...Burckhardt's and Damoiseau's Lunar Tables..."
4 pages.
1836: Woolhouse, W.S.B. "On Eclipses." 96 pages.
1837: Airy. G.B. "On the Calculation of. ..Perturbations..." 23 pages.
Woolhouse, W.S.B. "On the Determination of the Longitude..."
12 pages,
1839: Stratford. W.S. "On the Elements of the Orbit of Halley's Comet..."
79 pages.
1851: Adams, J.C. "On the Perturbations of Uranus." 29 pages.
1853-1914: Each contains the elements and ephemerides of a number (from 4 to
36) of minor planets.
1854: Challis, J. "On the Correction of a Longitude..." 23 pages.
1856: Encke, J.F. (trans. Airy, G.B.) "On a New Method of Computing the
Perturbations of Planets." 33 pages.
Adams, J.C. "On New Tables of the Moon's Parallax." 20 pages.
1862: "Comparison of Moon's Places by Burckhardt's Tables with Similar
Ones by Hansen's Tables." 2 pages.
1867: Breen, H. "Corrections... to... the [Tabulated] Values of the Moon's...
Parallax... 183 1-1839." 6 pages.
1874: "Predictions for the Transit of Venus." 6 pages.
1881: Adams, J.C. "Continuation of... Damoiseau's Tables of Jupiter's Satel-
lites." 9 pages.
1883 1922: Between about 1850 and 1900, a series of "Nautical Almanac Circu-
lars" was issued, mostly giving details and local predictions of total
eclipses.
1883-1922: Newcomb, S. "Corrections... to Hansen's Tables of the Moon." [These
contain longitude and latitude corrections only from 1883 to 1895,
those for right ascension and declination being included also from
1896.] 2-4 pages.
1897: "Approximate Places for 1900.0 of 834 Zodiacal Stars..." 11 pages.
688 EXPLANATORY SUPPLEMENT
1900: Downing, A.M.W. "Continuation of. . .Damoiseau's Tables of Jupiter's
Satellites." 7 pages.
1901-1906: "Corrections to the Apparent Places... to Obtain Apparent Places
Corresponding to the Struve-Peters Constants." Separate publica-
tions, 22 pages.
1907-1919: "Calculation of a Lunar Distance." 2 pages.
1907-1914: "Ephemerides for Physical Observations." 30 pages.
1915: "Some Constants and Formulae." 4 pages.
"Heliocentric [Co-ordinates] of. ..[Planets]." 68 pages.
1915: "Corrections to... 1532 Stars..." (Separate publication, by Pulkovo
Observatory), 308 pages.
1916: "Heliocentric [Co-ordinates].., of Venus." 82 pages.
1917: "Heliocentric [Co-ordinates]... of Mars." 82 pages.
1918: "Derivation of Quantities Contained in the Nautical Almanac." 23 pages
1920: "Ross's Corrections to... Places of Mars..." 16 pages.
1929: "Coordinates of the Sun for 1950.0 for 1928 and 1929." 32 pages.
1931: Fotheringham, J.K. "The Calendar." 14 pages.
"Derivation of Quantities..." 26 pages.
"Tables for Interpolation... by the End-Figure Process." 32 pages.
1935: Fotheringham, J.K. "The Calendar." 17 pages.
"Interpolation Tables." 14 pages.
1936: "Interpolation and Allied Tables." (Reprinted for separate sale) 48 pages
1938: "The Prediction and Reduction of Occultations." (Separate publica-
tion) 50 pages.
1938: "The Total Solar Eclipse of 1940 October 1." (Typescript) 19 pages.
1940: "Heliocentric Co-ordinates of Mercury." 4 pages.
"Corrections FK3 - Eichelberger." 6 pages.
1941: "Occultation Reduction Elements..." (Separate publication by Yale
University Observatory) 37 pages.
"Ephemeris of Pluto." 2 pages.
"Improved Lunar Ephemeris 1952-1959." (Separate publication as a
"Joint Supplement to The American Ephemeris and The (British)
Nautical Almanac") 435 pages.
14.62 The American Ephemeris
Appendix III of Volume VI of the Publications of the U.S. Naval Observatory, Sec-
ond Series, contains a list by William D. Horrigan of publications issued by the U.S.
Naval Observatory from 1845 to 1908. This lists the publications of the Depot of
Charts and Instruments, the annual volumes of the Washington Observations, the
Publications "Second Series," appendices to the Washington Observations, appen-
dices to the Publications "Second Series," meteorological observations, reports of
14 / RELATED PUBLICATIONS qqq
the Superintendent, sailing directions, wind and current charts, specifications relat-
ing to the new Naval Observatory, mathematical and astronomical tables, special
reports, observations, circulars, and reports on the U.S. Naval Astronomical Ex-
pedition to the Southern Hemisphere and the publications of the Transit of Venus
Commission.
Unlike most of the other national ephemerides. The American Ephemeris was
never a medium for the publication of technical articles. The volumes from 1855 to
1911, inclusive, contained an appendix, but ordinarily it comprised only the miscel-
laneous tables regularly included every year, and the list of fundamental constants
and tables used in preparing the ephemerides. In 1912, this list was transferred to
the beginning of the volume, leaving the section of miscellaneous tables at the end,
and the appendix was discontinued. Occasionally, appendices containing various
ephemerides have also been added to individual volumes; but very few technical
contributions were ever included.
However, separate supplements to The American Ephemeris have been issued
from time to time. Sometimes these were separate printings of material from the ap-
pendices but more often they were in addition to the contents of the volumes. Most
of them contain supplementary ephemeris data, especially for total solar eclipses,
but some have also been several important technical publications.
The appendices and supplements that are of interest for their technical content,
other than ephemeral data, are listed in Section 14.621. A list of the supplements
giving extended data and large-scale maps for total solar eclipses is given in Sec-
tion 14.622.
Prior to the establishment of the series of Astronomical Papers Prepared for
the Use of The American Ephemeris and Nautical Almanac (see Section 14.2 for
a complete list of contents), the principal tables constructed for the office were
printed, each as a single publication (a list of these is given in Section 14.623); but
many tables were prepared only in manuscript.
14.621 Appendices and Supplements The following are appendices and sup-
plements to The American Ephemeris:
1855: Chauvenet's tables for correcting lunar distances, with directions for
using the tables, and explanation of their construction. Pages 13-70
of the appendix.
1857: Chauvenet's tables for correcting lunar distances, with directions for
using the tables, and explanation of their construction. Pages 11-67.
Chauvenet, W. "Improved Method of Finding the Error and Rate of
a Chronometer by Equal Altitudes." Pages 69 94.
Walker, S.C. "Logarithms of the LeVerrier Coefficients of the Pertur-
bative Function of Planetary Motion." Pages 95 -117.
690
EXPLANATORY SUPPLEMENT
Pages 11-94 of this appendix were later reprinted as a separate publi-
cation. (Chauvenet, W. "New Method of Correcting Lunar Distances,
and Improved Method of Finding the Error and Rate of a Chronome-
ter by Equal Altitudes." 1866.)
1874: Coffin, J.H.C. "Tables for Finding the Latitude of a Place by Altitudes
of Polaris." (Supplement for 1874 1877).
This article, which included the formulas from which the tables were
calculated, with instructions for using the tables, and an illustrative
example, was also put into the Appendix (pages 25-33) in The Amer-
ican Ephemeris for 1877, the first volume in which these tables were
given.
A similar supplement for the years 1878-1881, inclusive, was sepa-
rately printed, and was also included in the Appendix in The Amer-
ican Ephemeris for each of these years. In the volume for 1882, these
tables were replaced by a simple one-page table which was retained
until it, in turn, was replaced by the table that was given throughout
the period 1912-1959.
1895: Newcomb, S. "The Elements of the Four Inner Planets and the Fun-
damental Constants of Astronomy." (Supplement for 1897.)
1945: "Tables of Sunrise, Sunset, and Twilight." (Supplement for 1946.)
1950: "Ephemeris of Pluto."
1954: "Improved Lunar Ephemeris 1952 1959." Published as a "Joint Sup-
plement to The American Ephemeris and The (British) Nautical Al-
manac."
14.622 Supplementary Publications for Solar Eclipses The series of supplements
to The American Ephemeris that were published for the occasions when total
eclipses of the Sun were visible in the United States began with a supplement
to the volume for 1869. Prior to that, pamphlets had been issued for the annular
eclipse of 1854 May 26 and the total eclipse of 1860 July 17.
For the total eclipse of 1869 August 7, a supplement was issued containing
predicted data, and also one containing suggestions for observing the eclipse. In
1885, a publication containing reports of observations of this eclipse was issued by
the Nautical Almanac Office.
Since the eclipse of 1869, supplements have been published for the total eclipses
that occurred on the following dates:
1878 July 29
1900 May 28
1918 June 8
1925 January 24
1932 August 31
1936 June 19
1940 October 1
1945 July 9
1947 May 20
1954 June 30
14 / RELATED PUBLICATIONS
691
In addition, data for two other eclipses were issued in U.S. Naval Observatory Circu-
lar no. 27 and U.S Naval Observatory Circular no. 78. The former gives information
on the annular eclipse of 1951 September 1, while the latter gives information on
the total eclipse of 1959 October 2.
14.623 Tables Prepared for The American Ephemeris and Nautical Almanac The
following tables were printed by the U.S. Naval Observatory:
Schubert, E.
Winlock, J.
Peirce, B.
Schubert, E.
Schubert, E.
Schubert, E.
Hill, G.W.
Todd, D.P.
"Tables of Melpomene." 1860.
"Almanac catalogue of zodiacal stars." 1864.
"Tables of Mercury." 1864.
"Tables of the Moon." 1st ed. 1853; 2nd ed. 1865.
"Tables of Eunomia." 1866.
"Tables of Harmonia." 1869.
"Tables of Parthenope." 1871.
"Tables to facilitate the reduction of places of the fixed stars." 1st
ed. 1869; 2nd ed. 1873.
"Tables of Venus." 1872 (on title page; cover has 1873).
"Continuation of de Damoiseau's tables of the satellites of Jupiter
to 1900." 1876.
1 4.63 Joint Supplements and Appendices
1968: The 1968 editions of The Astronomical Ephemeris and Nautical Alma-
nac and The American Ephemeris and Nautical Almanac contain The
Supplement to the A.E. 1968, "The Introduction of the IAU System
of Astronomical Constants." The supplement gives the values for the
astronomical constants, the theory of t he corrections, and the changes
in The Explanatory Supplement that result from these changes.
1968-1971: These editions of The Astronomical Ephemeris and Nautical Almanac
and The American Ephemeris and Nautical Almanac contain appen-
dices with the formulas and corrections for conversions to the IAU
system of astronomical constants. The ephemerides in the volume
were not based on the new constants, so the appendices were neces-
sary to give the corrections for the ephemerides to be on the system
of constants.
1984: The 1984 edition of The Astronomical Almanac contains Supplement
to the Astronomical Almanac 1984, "The Introduction of the Im-
proved IAU System of Astronomical Constants, Time Scales and
Reference Frame into The Astronomical Almanac." This supplement
gives the various resolutions involved and the resulting equations that
EXPLANATORY SUPPLEMENT
introduce the IAU (1976) System of Constants, the FK5 reference
frame on J2000.0, and the TDT and TDB time systems.
CHAPTER
Reference Data
Table 15.1
Fundamental Constants (1986 Recommended Values)
General Constants
Quantity Symbol Value Units Relative
uncertainty (ppm)
Universal Constants
speed of light
C
299792458
ms_1
(exact)
permeability of vacuum
Wl
4ttx 1Cr7
NA-2
12.566370614...
10-7NA-2
(exact)
permittivity of vacuum, 1 / tigC2
«S
8.854187817...
10-l£Fm-1
(exact)
Newtonian constant of gravitation
G
6.67259 (85)
10-11m3kg-1s-2
128
Planck constant
h
6.6260755 (40)
10-34Js
0.60
in electron volts, h I {e}
4.1356692 (12)
10-1sJs
0.60
hlZ-K
h
1 .05457266 (63)
10"34Js
0.60
in electron volts, h I {e}
6.5821220 (20)
10"16Js
0.60
Planck mass, (he! G)iiz
ITtp
2.17671 (14)
10-flkg
64
Planck length, h 1 mpc
lp
1.61605(10)
10-36m
64
Planck time, lp 1 c
tp
5.39056 (34)
10"44s
64
Electromagnetic Constants
elementary charge
e
1.60217733 (49)
10"19C
0.30
mfb
2.41798836 (72)
1014AJ"1
0.30
magnetic flux quantum, h 1 2e
$
2.06783461 (61)
10- 15 Wb
0.30
2ei h
4.8359767 (14)
1014HzV-1
0.30
quantized Hall conductance,
2a 1 fj,0c
3.87404613 (17)
10-5AV-1
0.045
7ab/e 15. 1 is continued on next page.
693
694
EXPLANATORY SUPPLEMENT
Table 15.1, continued
Fundamental Constants (1986 Recommended Values)
Quantity
Symbol
Value
1 1 nit o
unns
ntJidLivt;
iir\i^£irt',3inliii/ { rvnm\
unoertciiiuy i^jjpni/
DOnf rndyricL'tJifi at>< *-j"t?
MS
9.2740154 (31)
10-24JT-1
0.34
If! 8l6Cu0n VUH£>, /IfJ f \P/
5.78838263 (52)
10~5 eVT-1
0.089
in nercz, ^b / "
1.39962418 (42)
1010HzT"1
0.30
in kelvins, ^b / ^
0.6717099 (57)
KT-1
8.4
in wavenumbers, mb^c
46.686437 (14)
m-'T-1
0.30
nuclear magneton, e/i / 2mp
M/V
5.0507866 (17)
10"27JT-1
0.34
in electron volts, /i^/ {e}
3.15245166 (28)
eVT-1
0.089
in hertz, fin I h
7.6225914(23)
MHzT-1
0.30
in wavenumbers, 1 he
2.54262281 (77)
10-2m 1T"
1 0.30
in kelvins, m/v ' *
3.658246 (31)
10-4 KT"1
8.4
Atomic Constants
fine-structure constant
a
7.29735308 (33)
1Q-3
0.045
a 1
137.0359895 (61)
0.045
nyvjuci y i^tji laLdi il
ft.
10973731.534 (13)
m-1
0.0012
3.2898419499 (39)
1015 Hz
0 0012
ft»fjc
2.1798741 (13)
10-1fl J
0.60
in eV, Rmhci {e}
13.6056981 (40)
eV
0.30
DUI II 1 aulUo, u / + it.-.
ao
0,529177249 (24)
10_10m
0.045
Hartrpp pnprnu / A-xfnPin
Eh
4.3597482 (26)
aJ
0.60
— in eV Eh / \e\
27.2113961 (81)
eV
0.30
nnantiim of rim itatinn
3.63694807 (33)
10_4m2s-1
0.089
muon moiar mass
M(ii)
1,13428913 (17)
10-4kg/mol
0.15
muon magnetic moment
Mm
4.4904514 (15)
10-26 JT"'
0.33
fn Bohr magnetons,
Mm ' MB
4.84197097(71)
10-3
0.15
in niir*lp.ar mannfltnnc
111 1 IUOICjo.! 1 1 ld.yi ItrlvJI lt>,
Mm'M/V
8.8905981 (13)
0.15
mimn mannptir rnrvrnipnf aininmalv
y^(eh i 2m^) — 1
a„
0.001165923 (84)
7.2
miion nf-faptrir P(1 -u a 1
2.002331846(17)
0.085
muon -proton magnetic
moment ratio
Mm 'Mp
3.18334547 (47)
0.15
Droton mass
/Dp
1.6726231 (10)
10-27 kg
0.59
1.007276470(12)
u
0.012
In electronvolts, nripC2 / {e}
938.27231 (28)
MeV
0.30
proton-electron mass ratio
mp / me
1836.152701 (37)
0.020
proton-muon mass ratio
nip / m,,
8.8802444(13)
0.15
proton specific charge
e/ mp
95788309 (29)
C/kg
0,30
proton molar mass
M(P)
1.007276470 (12)
10~3kg/mol
0.012
15 / REFERENCE DATA
695
Table 15.1, continued
Fundamental Constants (1986 Recommended Values)
Quantity Symbol Value Units Relative
uncertainty (ppm
proton Compton wavelength,
A c,p
1.32141002 (12)
10-,5m
0.089
\c.p 1 2ir
2.10308937(19)
10-16m
0.089
proton magnetic moment
Mp
1.41060761 (47)
10-26 JT-1
0.34
in Bohr magnetons,
MpMB
1.521032202 (15)
10"3
0.010
in nuclear magnetons,
fJ.pl UN
2.792847386 (36)
0.013
diamagnetic shielding
correction for protons
in pure water,
10-6
spherical sample, 25°C
25.689 (15)
shielded proton moment
(H20. sph., 25°C)
1.41057314(47)
10-26 JT-1
0.34
in Bohr magnetons,
Hp' MS
1.520993129 (17)
10-3
0.011
in nuclear magnetons,
Mp f MN
2.792847386 (63)
0.022
proton gyromagnetic ratio
fp
267522128 (81)
0.30
Physico-Chemical Constants
Avogadro constant
6.0221367 (36)
1023 mol-1
0.59
atomic (unified) mass unit,
atomic mass constant
1u = wu = 1 / 12m(12C)
ftlu
1.6605402 (10)
10"27kg
0,59
in electron volts, rtiuC2 / {e}
931.49432 (28)
MeV
0.30
Faraday constant
F
96485.309 (29)
Cmor1
0.30
molar Planck constant
NAh
3.99031323 (36)
10-10 Jsmol-1
0.089
NAhc
0.11962658(11)
Jm mol-1
0.089
gas constant
R
8.314510 (70)
J mol-1 K-1
8.4
Boltzmann constant, R / WA
k
1.380658 (12)
10-23 JK-1
8.4
in electron volts,
k/{e}
8.617384 (72)
10-s eV/K
8.5
{e}/ k
11604.45 (10)
K/eV
8.5
molar volume (idea! gas)
8.4
f = 273?15K;p= 101325 Pa
vm
22.41410(19)
Umol
Loschmidt number, WA / v"m
no
2.686773 (23)
1025m^3
8.4
T = 273°15K; p= 100kPa
Vm
22.71108 (19)
Lymol
8.4
Loschmidt number, WA / Vm
no
2.651629 (22)
lO2^"3
8.4
Stefan-Boltzmann constant
a
5.67051 (19)
10-8 Wm-aK~4
34.
first radiation constant, 2-trhc2
Cl
3.7417749 (22)
10-16 Wm2
0.60
second radiation constant, he I k
C2
0.01438769 (12)
mK
8.4
Wein displacement law constant.
b= AmaxT= C2/4.96511423 .
b
2.897756 (24)
10-3mK
8.4
Note: The digits in parentheses are the one standard-deviation uncertainty in the last digits of the given vaiue.
696
EXPLANATORY SUPPLEMENT
Table 15.2
IAU (1976) System of Astronomical Constants
Units: • The units meter (m), kilogram (kg), and second (s) are the units of length, mass, and time
in the International System of Units (SI).
• The astronomical unit of time is a time interval of one day (D) of 86400 seconds. An interval
of 36525 days is one Julian century.
• The astronomical unit of mass is the mass of the Sun (S).
• The astronomical unit of length is that length (A) for which the Gaussion gravitational constant
(k) takes the value 0.017202098 95 when the units of measurement are the astronomical
units of length, mass, and time. The dimensions of k2 are those of the constant of gravitation
(G), i.e., L3M~1 T~2. The term "unit distance" is also used for the length A.
• In the preparation of the ephemerides and the fitting of the ephemerides to all the obser-
vational data available, it was necessary to modify some of the constants and planetary
masses. The modified values of the constants are indicated in brackets following the (1976)
System values.
Defining constants:
1 . Gaussian gravitational constant
2. Speed of light
Primary constants:
3. Light-time for unit distance
4. Equatorial radius for Earth
IUGG value
5. Dynamical form-factor for Earth
6. Geocentric gravitational constant
7. Constant of gravitation
8. Ratio of mass of Moon to that of Earth
9. General precession in longitude, per
Julian century, at standard epoch
J2000.0
1 0. Obliquity of the ecliptic, at standard epoch
J2000.0
Derived constants:
11. Constant of nutation, at standard epoch
J2000.0
12. Unit distance
13. Soiar parallax
14. Constant of aberration, for standard
epoch J2000.0
1 5. Flattening factor for the Earth
16. Heliocentric gravitational constant
fc = 0.01 720209895
c=299792458 ms"1
ta= 499.004782 s
[499.00478370...]
ae =63781 40 m
ae= 63781 36 m
J2= 0.001 082626
G£=3. 986005 x 1014m3s-2
[3.98600448 ... x 1014]
G = 6.672 x 10" 11 m3 kg"1 S"2
[6. 67259 x 10"11]
H = 0.0 1230002
[0.012300034]
p = 5029:'0966
e= 23° 26' 21 "448 [23° 26'2i:'4119]
W = 9"2025
Cta = 4=1. 49597870 x 1011 m
[1.4959787066 x 1011]
arcsin (ae / A) = ?r0 =8?7941 48
[8.794144]
k = 20'.'49552
/ = 0.00335281
= 1 / 298. 257
A3k2 I D2 = GS=1. 32712438 x 1020 m3 S"2
[1.32712440 ... x 1020]
15 / REFERENCE DATA
697
Table 15.2, continued
IAU (1976) System of Astronomical Constants
1 7. Ratio of mass of Sun to that of the Earth
18.
19.
20.
Ratio of mass of Sun to that of
Earth + Moon
Mass of the Sun
System of planetary masses
Ratios of mass of Sun to masses of the planets
{GS)I{GE) = SIE = 332946.0
[332946.038...]
(SlE)l{\ +/1)= 328900. 5
[328900.55]
(GS)/ G= 5=1.9891 x 1030kg
Mercury
Venus
Earth + Moon
Mars
6023600
408523.5
328900.5 [0.55]
3098710
Jupiter
Saturn
Uranus
Neptune
Pluto
1047.355
3498.5
22869
19314
3000000
[1047.350]
[3498.0]
[22960]
[130000000]
Other Quantities for Use in the Preparation of Ephemerides
It is recommended that the values given in the following list should normally be used in the nreDaration
of new ephemerides. H K
21 . Masses of minor planets
Minor pfanel Mass in solar mass
22.
(1) Ceres
(2) Pallas
(4) Vesta
Masses of satellites
5.9 x 10-10
1.1 x 10-10 [1.0814x 10"
1.2 x10-10 [1.3787x10"
23.
24.
Planet
Satellite
Satellite/Planet
Jupiter
lo
4.70x 10-5
Europa
2.56x 10^5
Ganymede
7.84 x 10-5
Callisto
5.6 x10-5
Saturn
Titan
2.41 x 10-4
Neptune
Triton
2 x 10~3
Equatorial radii in km
Mercury
2439
Jupiter 71398
Pluto 2500
Venus
6052
Saturn 60000
Earth
6378.140
Uranus 25400
Moon 1 738
Mars
3397.2
Neptune 24300
Sun 696000
Gravity fields of planets
Planet
J2
Ja
Earth
+0.00108263 -0.254 x 1Q-5
-0.161 x10"5
Mars
+0.001964
+0.36x 10-4
Jupiter
+0.01475
-0.58 x 10"3
Saturn
+0.01645
-0. 10 x 10"2
Uranus
+0.012
Neptune
+0.004
(Mars: C22
= -0.000055, = +0.000031, S^ = +0.000026)
25.
Gravity field of the Moon
7 = (S- A)/ C = 0.0002278 C/AOT^O.392
0=(C - A)! 6 = 0.0006313 / = 5552"7 = 1°32'32:'7
C20 = - 0. 0002027 C30 = - 0. 000006 C32 = +0. 0000048
= +0. 0000223 Cg, = +0. 000029 S32 = +0. 000001 7
§3, = +0. 000004 C33 = +0. 000001 8
S33 = -0.000001
698
EXPLANATORY SUPPLEMENT
Table 15.3
Time and Standard Epochs
1 day = 24 hours = 1440 minutes = 86400 seconds
1 Julian year = 365.25 days = 8766 hours = 525960 minutes = 31557600 seconds
Length of the Year at 1990
d
d
h
m
s
Tropical (equinox to equinox)
365.2421897
365
05
48
45.19
Sidereal (fixed star to fixed star)
365.25636
365
06
09
10
Anomalistic (perihelion to perihelion)
365.25964
365
06
13
53
Eclipse (Moon's node to Moon's node)
346.62005
346
14
52
52
Gaussian (Kepler's law for a = 1)
365.25690
365
06
09
56
Julian
365.25
365
06
00
00
Length of the Month
d
d
h
m
s
Synodic (new moon to new moon)
29.53059
29
12
44
03
Tropical (equinox to equinox)
27.32158
27
07
43
05
Sidereal (fixed star to fixed star)
27.32166
27
07
43
12
Anomalistic (perigee to perigee)
27.55455
27
13
18
33
Draconic (node to node)
27.21222
27
05
05
36
Length of the Day
1d of mean solar time = 1 ?00273790935 of mean sidereal time
_ 24h03m56?555368 of mean sidereal time
= 86636.555368 mean sidereal seconds
1" of mean sidereal time s 0^99726956633 of mean solar time
= 23h56m04f 09054 of mean solar time
= 86164.09054 mean solar seconds
15 / REFERENCE DATA
699
Table 15.3, continued
Time and Standard Epochs
Standard Epochs
Julian Year epochs
J.Y.
Julian date
J 1900.0
2415020.0
J 1950.0
2433282.5
J 2000.0
2451545.0
J 2050.0
2469807.5
J 21 00.0
2488070.0
1900
January 0.5 =
JD 2415020.0
1925
January 0.5 =
JD 2424151.0
1950
January 0.5 a
JD 2433282.0
2000
January 0.5 =
JD 2451544.0
2050
January 0.5 =
JD 2469807.0
2100
January 0.5 =
JD 2488069.0
Beginning of
Besselian year
B.Y.
Julian Date
B.Y. Julian Date
B1 850,0
2396758.203
B2000.0 2451544.533
B1 900.0
2415020.313
B2025.0 2460675.588
B1 950.0
2433282.423
B2050.0 2469806.643
B 1975.0
2442413.478
B2100.0 2488068.753
Table 15.4
Sun, Earth, and Moon
IAU System
Best Estimate
Sun
Radius
Semidiameter at mean distance
Mass
Mean density
Surface gravity
Inclination of solar equator to
ecliptic
Longitude of ascending node (7" in
centuries from J2000.0)
Period of synodic rotation (<j> =
latitude)
Period of sidereal rotation adopted
for heliographic longitudes
Motion relative to near stars
6.96 x 108m
15'59:'63 = 959:'63
1.9891 x 1030kg
1.41 gem-3
2.74 x K^ms-2 = 27.9g
7° 15'
75° 46' +84' T
26. 90 + 5. 2 sin2 <p days
25.38 days
apex: q = 271 ° 8 = +30°
speed: 1 . 94 x 1 04 m/s = 0
Figure and Gravity Field of the Earth
Equatorial radius
Dynamical form factor for Earth
Flattening
Polar radius
Mass of the Earth
Mean density
Normal gravity (g)
Geocentric gravitational constant
2.8x 10"
a = 6378140 m 6378136 m
jfe = 0. 001 08263 0. 001 082626 ■
/ = 1 / 298. 257
b = 6356755 m 6356752 m
5. 9742 x 1 024 kg 5 . 9742 x 1 024 kg
5.52 g/cm3
9.80621 - 0.02593 cos 2<p + 0.00003 cos 4^ m/s2
3. 986005 x 1014m3s_a 3.98600440 x 1014m3s-2
1° of latitude
1 0 of longitude
Geodetic latitude (<f>)
latitude (</>')
For a point on the spheroid of the IAU System at geodetic latitude (<j>):
110.575 + 1.110sin2<f.km
{111. 320 + 0. 373 sin2 <j>) cos <j> km
692:74 sin 2<p - 1 "1 6 sin 4<£
geocentric
Solar parallax
Constant of aberration (J2000.0)
Light-time for 1 au
1 astronomical unit of length
Mass ratio — Sun/Earth
Sun/Earth + Moon
Moon/Earth
Mean eccentricity
Mean obliquity of the ecliptic
Annual rate of rotation on the
ecliptic
Mean distance of Earth from Sun
Mean orbital speed
Mean centripetal acceleration
Orbit of the Earth
8'.794148
20'.'49552
499.004782 s
1.49597870 x 101,m
332946.0
328900.5
0.0123002
0.016708617
23°26'21°448
0:'4704
1.000001 01 78 AU
29.7859 km/s
0.00594 ms-a
8:794144
499.00478353 s
1.4959787066 x 1011m
332946.045
328900.55
0.012300034
23°26'21:4119
1.000001 05726665 AU
29.784766966 km/s
5.9301134387 x 10"3ms
15 / REFERENCE DATA
Table 15.4, continued
Sun, Earth, and Moon
IAU System
Best Estimate
Orbit of Moon about the Earth
Sidereal mean motion of Moon
Mean distance of Moon from Earth
Equatorial horizontal parallax
at mean distance
Mean distance of center of Earth
from Earth-Moon ban/center
Mean eccentricity
Mean inclination to ecliptic
Mean inclination to lunar equator
Limits of geocentric declination
Saros = 223 lunations = 19 passages of Sun through node = 6585 1/3 days
Period of revolution of node 6793d
Period of revolution of perigee 3232s
Mean orbital speed 1023 ms-1 = 0.000591 Au/d
Mean centripetal acceleration 0.00272 ms~2 = 0. 0003 g
2.661699489 x 10-6rads-1
3.844 x 105km
60.27 Earth radii
0.002570 AU
57'02'.'608
3422'.'608
4.671 x 103km
0.05490
5° 145396
6°41'
± 29°
Period with respect to fixed stars
in mean sidereal time
in mean solar time
Rate of rotation
Rotation of the Earth
24h00",00s0084
23h56m04f0989
15'.'04106717866910s-1
7.29211510 x 10-5rads-'
Annual rates of precession (T in
centuries from J2000.0)
general precession in longitude
lunisolar precession in longitude
planetary precession
Mean radius
Semi-diameter at mean distance
Mass
Mean density
Surface gravity
50:290966 + 0:0222226 T
50'.'387784 + 0:0049263 7"
-0.0188623 - 0:0476128 T
Moon
1738 km
15'32:'6
7. 3483 x lO^kg
3.34 gem-3
1.62ms-2 = 0.1 7g
702
EXPLANATORY SUPPLEMENT
15 / REFERENCE DATA
703
o> t: to
> CO P
. lli ^
ra^: i
S o o
o o
■o
£
Q.
co
0)
o
c
0)
0) —
DC
75 cm
- (0
c "
(0 E
§ °
o lj —
■5 Q- 1
E
B
I
a?
CD
Mi cr
ui o
5 a
o
3 1
cr ™
LU DC
in
i e
0)
Q
C
co
m
E
CO
w 1
^ 1
i!
CM
CM
Cvl
CJ>
cn
cn
CM
cvl
CM
1^
CO
CO
CD
CD
C\J
OJ
CM
CM
CO
CO
co
CO
O
o
o
o
5
5
o
o
o
o
o
o
d
d
d
d
II Mill
in
co
1
o
o
1
'CO
1
CO
1
1
1
1
1
1
o
5
o
o
5
o
CD
cc
CO
CO
CO
CO
as
s
CD
CD-
co
■CO
CO
CO
CM
f-
CO
CO
CM
CD
Gj
CO
CM
CO
CO
5
r-
CCD
r-
s
r-
r-
CD
CO
CM
LO
LO
uo
LO
■=3-
CO
CO
CD
r-
LO
o
CM
CM
CM
CM
CM
CM
CM
CM
CO
CD
O0
CO
CO
CO
03
00
00
CO
00
CO
f^-
CO
■*
CD
d
CD
CO
CD
CD
CD
CD
CD
CD
cn
CS
cn
CD
CD
o
CD
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CO
CM
LO
LO
LO
CO
LO
CD
CO
CO
r-
r-
O
o
O
O
O
CO
CO
ai
CD
CO
CO
CO
m
CO
CD
CO
CO
&
CO
O
CD
r-
(0
CM
CO
CM
CM
CO
CM
LO
CO
CO
CO
CO
5
CO
CO
CC'
CO
co
CO
r-
S
o
CO
s
— ' CD
°
co en
O <
LO
CD
O) j-^ CD
CO ^ *~
i s «
§5 1
O (TJ CO
— "5
i
CD
E ~
< r»- _
co co cn
S » §
o 0- =!
C/> O <
3
< *
2 o>
_ o
(0 s- o co *; co
c £ a eg 5 co
o >i co co co t— o
3g co »- ^ co
<3 X cu o oi £ »
a « S $ ^ £■
O O CD LU <
■p
01
3
c
+-
c
o
u
in
LO
XI
t—
g
1
S
CO
CO
CO
CM
T
CO CO
CO
LO
00 r-.
9
1
O
T
LO CM
^ 1
CO
1
CM CM
CM t^-
1 m
1 I
•tf CM (O rt CO
CM CO CO T-
0
CD
CO
T
CO
in
LO
CD
CO
h-
iO
LO
10
q
in
OJ
CO
CM
CD
■A
w
UO
CM
CO
LO
CO
CO
h-
S3
s
CM
IO
5
LO
5
CD
CO
CM
n
CM
03
10
cn
LO
to
CM
5
LO
LO
CM
T
CO
CO
LO
cv
^«
cn
O
CM
OC
CD
CO
CM
10
OJ
CO
LO
OJ
LO
LO
CO
I I
o
E
IB
a
0!
a
ctj
O E
c'|^ 8 g m
0>nia>CB:V,G):cjicc
cn
cvj co
_ o
CO CO CO
CO CD
— CO
" .3 10 -S I - n s -
<
™ 2 g JS " E
LliU<£CQ!i:COiIJ
- n I <»
z <
(0
w o
,,, 01 (J)
5 o ^ 5 1:
^ .9 o 3 o
O o_ a 0. to
i
O
O
o
g
0
T3
0) C3
■a S
C3 CO
^_ *£ nl ^ rr CO
w o = :
CD CO » ;
11 no
O a1 d'<5>-6-";e-
EXPLANATORY
Table 15.6
planets: Mean Elements
(For epoch J2000.0 = JD2451 545.0 = 2000 January 1.5)
Planet
Inclination (/')
Eccentricity (e)
Mercury
7°nn' 1 yqso^i
0.2056317524914
Venus
3°23'40"Q7828
0.0067718819142
Earth
0.0
0.0167086171540
Mars
1 ° EfV RQT11 RT?
Jupiter
1°18' 11:77079
0.0484948512199
Saturn
2°29' 19196115
0.0555086217172
Uranus
0°46' 23:50621
0.0462958985125
Neptune
1°46' 11:82795
0.0089880948652
Pluto
17°08'31:8
0.249050
Mean Longitude
Mean Longitude Mean Longitude
of Node (f?)
of Perihelion fro) at Epoch (L)
Mercury
*ttj 1 J D 1 1
77° 27' 22'.'02855 252° 1 5' 03:25985
Venus
76° 40' 47:71 268
1 31 0 33' 49'.'34607 1 81 c 58' 47:28304
Earth
0.0
102° 56' 14:45310 100°27'59:21464
Mars
49° 33' 29: 13554
336° 03' 36'.'84233 355° 25' 59:78866
Jupiter
100°27'51:'98631
14° 19' 52'.'71326 34°21'05:34211
Saturn
11 3° 39' 55:88533
93°03 24"43421 50° 04' 38:89695
Uranus
74°00'21:41002
173°00'18US7320 31 4° 03' 18:01840
Neptune
131°47' 02:60528
48°07'25-28581 304° 20' 55: 19574
Pluto
110°17'49r7
224° 08' 05:5 238°44'38r2
Mean Distance (au)
Mean Distance (1Q11 m)
Msrcury
0.3870983098
0.579090830
Venus
0.7233298200
1 .08208601
Earth
1.0000010178
1 ,49598023
Mars
1.5236793419
2,27939186
Jupiter
5.2026031913
7.78298361
Saturn
9.5549095957
14,29394133
Uranus
19.2184460618
28.75038615
Neptune
30,1103868694
45.04449769
Pluto
39.544674
59.157990
Sidereal Synodic Mean Daily Orbital
Period (Julian years) Period (d) Motion (n) Velocity (km/s)
o
Mercury
0.24084445
1 15.8775
4? 09237706
47.8725
Venus
0.61518257
583.9214
1°60216874
35.0214
Earth
0.99997862
0? 98564736
29.7859
Mars
1.88071105
779.9361
0° 52407 109
24.1309
Jupiter
1 1 .85652502
398.8840
0° 0831 2944
13.0697
Saturn
29.42351935
378.0919
0° 03349791
9.6724
Uranus
83.74740682
369.6560
0? 01 176904
6.8352
Neptune
163.7232045
367.4867
0? 006020076
5.4778
Pluto
248.0208
366.7207
0? 003973966
4.7490
Note: The elements for Pluto are based on a tit over one century. Trie values for the other
planets are based on an evaluation of the expressions for the mean elements evaluated at
the epoch.
Table 15.7
Planets: Rotational Data
(North Pole of Rotation and the Prime Meridian)
Sun
ao
= 286.° 13
*o
= 63°87
W
= 84°10
+ 1 4? 1 844000 d
Mercury
ao
= 281.01
-0.003 7"
*o
= 61 . 45
- 0. 005 r
W
= 329. 71
+ 6.1385025 d*
Venus
a0
= 272. 72
60
= 67.15
W
= 160.26
- 1.481 3596 d
Earth
a0
= 0,00
-0.641 T
60
= 90. 00
-0.557T
W
= 190,16
+ 360. 9856235 d*
Mars
ao
= 317.681
0.108 7
60
= 52.886
- 0. 061 7
W
= 1 76. 868 + 350. 89 1 9830 d*
Jupiter
a0
= 268. 05
- 0. 009 T
60
= 64. 49
+ 0. 003 7
w
= 284. 95
+ 870.5360000 d§
Saturn
"0
= 40. 58
- 0. 036 7
*0
= 83. 54
- 0. 004 7
W
= 38. 90
+ 810.7939024 d§
Uranus
= 257. 43
bo
= -15.10
= 203. 81
-501.1600928 S
rwpiune
= 299.36
+ 0.70 sin N
to-.
= 43. 46
- 0.51 cos N
W:
= 253.18
+ 536. 3128492 d-
N:
= 359. 28
+ 54. 308 7
Pluto
a0 ■■
= 313.02
S0-
= 9. 09
W-.
= 236. 77
-56. 36231 95 d
a0, 60 are standard equatorial coordinates with equinox J2000 at epoch J2000. Approximate coordinates of
the north pole of the invariable plane are «0 = 273°85, b0 = 66?99.
W = location of the prime meridian measured along the planet's equator in an easterly direction with respect
to Ihe planet's north pole from the node (located at right ascension 90° + n0) of the planet's equator
on the standard equator. If W increases with time, the planet has direct rotation and if W decreases
with time, rotation is said to be retrograde,
f = interval in Julian centuries (of 36525 days) from the standard epoch,
d = interval in days tram the standard epoch.
The standard epoch is 2000 January 1.5, i.e., JD 2451545.0 TDB.
Notes: * The 20° meridian is defined by the crater Hun Kal.
* The expression for W might be in error by as much as 0.2 because of uncertainty in the length of the UT
day and the TDT - UT on 2000 January 1 .
* The 0° meridian ts defined by the crater Airy-O.
5 The equations for W for Jupiter, Saturn, and Uranus refer to the rotation of their magnetic fields (System
III). On Jupiter, System I (Wi = 67? 1 +877° 900 d) refers to the mean atmospheric equatorial rotation:
System II (IVM = 43°3 + 870.°270 d) refers to the mean atmospheric rotation north of the south component
of the north equatorial bell, and south of the north component of the south equatorial bell.
706
EXPLANATORY SUPPLEMENT
15 / REFERENCE DATA
707
O
m
>■
ra
C s
a c
o «! /
£ ,
Ol
| E
S I
ra cn
« E
fa S
ui E
b a
< 5
cc — '
E
I
111 II
I I
m rt t-
OJ. m
tri io lo
5t .
ID CM |
d d
co in
cm §
<j! S ffi
2 > lil
1 7
o <o m m
rj o) n *t
o -i- ^
3 £3
CO Ol CO
co CO »-
O ID OJ
O O O
O O O O
fa
N m cd
O CM O O
o in f\l ui
6 6 t (O
00 S CM (D
co o a> cd
CO ^ CO
1-i CO r- O
N CO
SB
o <S
o o
CO m
CO CO
01 CO
CO LO
C CO t
O « = 3
II = »
e. 2 -5 to
I I I
la I I III
O T- 1- T-
o
CO -tf CO
CD «* cm
h- <D Cd
tD O)
id 'j n cSj
CD CO t-
n w o
in io irt
LO h- i-
m t}- -i—
eg cv
u> in
OJ CO
y) r- o
If I
3 2 0.
_ T>
i 3
If
B 3 >;
E ST O
U I" O
g 0 5
■ O CQ O
I I
t co
I
CO CO
m -*
CM
I
S5
0j CO
13 UJ
co cd co co
Tf m in
Q O i-
i I i «
O r~
OJ i OJ c\i
t— CM O 00
CM lO CO
d t- ai cd
+ 111
o o o o
CO OJ OJ CO
ID r r N
cd in CO <D
OJ CM
To jB
in >> (0
" &
CO " CO
(O Ifl t ^
co cd in m
^ «i n s
OJ CVJ |f- CO
co o «< 5
K d O
CM
O 10 .IS 3
o s e- S
£2
en r~
t- co o
cd ^
I I I
in >t
d d
CD CD
co m
d o-
05 cm co
as
Q
a.
1
3
_C
ra
c
:al
CO
ysi'
cd
Ph
T—
CO
<D
jQ
cn
Ian
r- tL
co
Q re
"5
01 Q)
CO" ^ J,
B
O
a
o
» £ i 1
^ x
cr
> £
Ft, Q
(9 <
ra o
LU
co ?
i "
P C
10 40 r*>
E « to
rd (u xj
^ = c
"C ra o
<U JZ CD
X *- Ui
'-n ® rz
C (A o W
:= ^ O Q- O
T3 K — m O
ScofS
™ t3 °
O
10
a.* 8
a> ,s g « ^
E
o
o
■ a a »
E .a i | f
m ^ 5 in °
a 3 Z - £
™ £ o £
o a; ra g jj
lifts
r ^ ^ f» f
U > CZ Q_
« S -2 ^ ffl h
c cn a c x: m
£ CD o 2 ^"
o> £ *" To d ^
-5 o
03
c «
ta ra
OJ ^
a.
& E
s 8
ra S
CL —
S ™
ca co
UJ a
<•> I
^- ca
I 1
-a cr
E o
° Jo
i ra
9 E
to 3
Q c
io
co cr
£ S
o e
co co
T3 £
ra i.
E
S 1
ca
E «
£ E
5 8
a» co
^ >> c
«; sp
ra ra JS co
i * s 1
^3 £
B (0
^ XJ
is £ ra
- £ £ * ~ o *
- - ~ o ° ^ -
E
O CO
Ol
c c
'i: n
> t
co co ai
r c r
l- j- h-
t— cvi co
(— m ^ — ;
" O CD ro
~ S &
S ra a
ra
^ i9 ^
ra o co
0> 0)
>• o o> o> a) cn ra
CO CD
3 s
cr CO
CO «-
nj Co
a io
■D CO
8 s
ill
" R »
£ S <u
o ra J=
e.e:
o p «
CJ ' o
H £ D.
« o "
ST- m ^
O (0
CD
C.Q.£
TJ 2
E £ w
a> -a -g
n c co
J— CO CO
EXPLANATORY SUPPLEMENT
nCDEDtOOtDlDtO
CT3 00 id ori c\i CM d
t- to tj- i- i-
CO
in lo lo cd oj
Q C\J C\J CP -i- LQ
LO CD CJ O O O
ffl ifi S CO t-
co i-
I
O 03
O O O O O N '3"
1- o ■>- o o o o
o o o o o o o
m r--
g £
i- f-- co
NcomNiocoNcn
cmcocm*-cM'-i-o
W O O O O)
LO O CM O 1—
_ ^ O OJ i- CD
CM O Q O O CM O
ooooooooooooooooo
o o
d b
O O CP t-
00 (O
CM OJ
do h
CM CD O
O T- O
o o o o o o a
CM OJ to o 2 CO
LOOCQ-tfOOOi-rO
i- o s
CO CO CO CO
CO t— ^ N
o o o o o
O O CM O O
LO h-
CO CO
CM CM
CO CM
cm -r
CM CM
tj- cm en co lo cd r^h- t-
05 OJ CM CM CO CO CD CM CM
OCMt-i-i-CM CM co lo cm
■>- CM ■>-
CD LO LO
m 0) i-
CO CM
t- OJ O ID O
CO LO O *-
■i— -i- CO CO CD ^
CO T- -&
CD CD
00 CM
CM t— to
CO CO CM
CD 1-
C0 CO LO
r~~ o ^ o
h- i- CM CO CJ5
CO CO to f s
t- r- Ul O i- CM CO
CB t- O) CD (D OJ
CO LO LO CO CD CO LO CM CM
NiflT-CfittOCDCCCCCMLlirN
^rtNcdddo)U5comcAiT-cci
i- LOIOCOLOLOOJCOCO
CM(MSNC\JCDCDCM
CM
CO LO O
*T LO co
CO 00 t-
O T- l>
CD CO CM i-
lO CM N « N
t CO (M O
h- oi en *t
CO CM CM CD CO
CM CO
-t CO CO
■ CO CO lO
03 CM
S o
oi in
ID CD o
f0 ^ ^ N P)
1X3 Ol CM (0
OOOOi-i-Pd^f
£ £
5 0. Q
E 2 £ 5
_ d of n
o ILi CD O
E E «
< X LJJ
OJ CD 03
c n
<
^ CD
Q 03
ID >
_J F <
£ = s a- £
LU F Q X F X
= __. = = > ^> 5
= = X - = = >>>
= S£>>>>SX
CO ^
111 s
15 / REFERENCE DATA
709
en lo
o o
o o
CO CTJ
CM CO
CO CD □ ^ CO LO O
cdrtcMVroomoistficocomNT-d d cd i- co t- to
T-lOT-CMLO^tCMOCMCDCO CM LO CO CO
LD^CJCMCMCMCMt'i- CO LO ^ CM t-
_ in
to 7 o i-'tn ^■i-Ninom
co a o nro^^ojt-i-wowrT-nonrtcostyoowLii,-
6 d o o o o o d d o d d d o o o on i^^dddddcD
UD CM Cft
CO
^ O CM CO N O CM <4
O W ^ COLOCMOCM-^Ot-t-t-t-t-i-t-^O t- t- -A ,- t-
OOO OOOOOOi— OOOOOOOOO LOOOOOOO o
poo OOOOOOOOOOOOOOOOKOOOOOOO
odd ddddddddddddddddddddddd d
v vvvvvvvv
V V V V
CM O CO O CO
WOCDCDSifiOcOCNJOrWOJNCDNCOCOlDCIl^CprCCl
^^©u)CnNmqqa)Lr)p)NS\-N(Dcooci)WON co to to o o cm cd
i-N^^NaT-rti-tflirictioimricj^oi^cd ri s ri w cm d cd cri
L0NCI)0)OC0^C0CS(0WC0W^iDi0CDC0t0C0l0h-(SLn t- t- N in CD m 't i-
t-cocmcmt-t— t— 1— t— cm^lo-t- n lo t-
lo
LO
"O
F) CM N O) C\J
or^i-oooicoo>iv.o^(ocDa>LO-cM^- <o lo m io m
^ssscONcoosMocoaimainm cd en co in fl- in t- lo
wocoa)CiOLnooT-ccwTto^«iO(OO^^tncocoT-co^icNmooj
^cD^^^-ncolno^^^c^Lo^t««wncow^co^O'^'-'TCo■^^l^l^
ffiCOCOCOO^CMNCM^OCOT-OSrtcO'NCni-inCMCDN CO Ol i— CO CM LO CM CO
^^■cqo3(oacOlnln^^Tl^^Rn^^^^lotn(D^ffl t- cm co co in t- cq
dcM^^ddddcM^cddi-^ddodobbbddio d d d d d d -<- to
*- CO
CO
"5 Q3 R w
E _
a. cd
w ™ ra
a oi ?v *- *- iu co v. tz *- jj
LUIFO<Q-Llll<DhO
cd ca ^ _c
s S B S I S I « I 11
UOcaOQ^CLCEmQ-F zz
F Q
s 8 1 §
55 £ o
_--_;>>>>
XXXXXXXX —
= = 2>>>>xx
_ - = > >
X X X X X
= ^ > > > > —
10
EXPLANATORY SUPPLEMENT
CQ
I
00 O CM O iO
i- m in tq in
o t- o o o
O ffl 5
CO CJ CO
odd
oooooooo
I
EG
000-»-0 0 0*-QOO
r*. h<- ^ h-
d d d d
S CO r CO CO, 0Q N O
r-- f-v. k c\j h"-; i> £>
dodo i- odd
*t O CM O) r in 3;
s^^qwtqcDi-cq
co cn
^ S CM i- i-
d oi cd *
i- en m r 3) m tj-
CMtocd'Xi'tqq^T-
d-i-oii-^-r^co-^r^ocj
+ + + I 1 1 1 + +
fy, CO CO CO o>
oniDNcQC\|ino^coni-©coT-c\|cpinK
*7— t*~ ^™ t~~ t- i^— + T"~ T~~ ^- + ^ + T I "P T T"
+ + + + + + + + +
cm o in co co
•f CM q p q
odd dodo* odd
io m in
o o q q q
d d d d -r^
OlNSWCOrvjOCO
o odooc:*!—
°- o
t ID
d d
CO CO to 0) 05 03
ffl
3
co 5
DC i
o
01 T-
CO CO
co in co
X
CD o o o m ID
o o o
CO i-
r- i- t-
S S:
n n in ifl
, I I I o o
o o o o i i
^~ "~ o o .
X X X X o o
CO CM Q CO X X
in x a) in a) (O „ „
n * (
CM CD
o I
o U. oi
CM X ^!
:o o a a *~ oo co
p ^ .JS I ^ ^ m co i co to„
I I I I o ° I I I I I o I On I 2
OOO0*-| OOOOOt-Ot- 0|
°£xxx^*°xxxxx*x£?x5
x x
x m ^ X X X
in cm o j<i-ioonrtffl'l-n x co x
io coK^*-ddodd^ddco^^T-^cMcocor^
cn
o
i e c =
'id A D o is
□ S 111 (3 O
„ iu _
E S J= =
C CO (D 03
CL C/3 Jj O < J I —
"O
(fl .35 ™
[/J (tf (U >■■
e § £
*= 9
b
=3 -£ <fl
> > > > 5 X
xxxxxx_:= = >>>>>*
15 / REFERENCE DATA
711
CQ
I
3
CM
d
l
DO
o o o o
in m co co co ca co
^ *t co co dTtco'dcMcg^-^cvjccid co aD'^cocj^cvidco
m o cm co
^ t °? ^ *i-ooiio*^n«icofflnsnio«iooi-onDmcoiii
m 5 co t» o co co i-cMT-^co^^dfflcscDCDoiaiK'WTfdair^KNifl'd
+ + T + + + + + + + + +T— T-T— l+T- ++ +
I o
o a>
k ?t co r^-rf- f--r^h--r^-j^-[%--r^r--h-{v^rv to to to to to to
sjN^tpfliqqui n r cm cm cmoooooooooo rv. ttooooooin
oooooooo d ddo ddddddddddd d dddddddd
to e/> c/3 co co
T3 e
to 5
O LO CO i- O II)
lO t- t- r- ^ CO
X X X X XX
o co o o to
cd i- i- t- v m
en cd o cm o ir> a
o r- co c?) co^tenmT-i-swVNnNCONOJO^CB
LD iflNNCyT-T-CMCOCM^lflMrtKT- t- W 9 S S
CO CO
O 1- CD
i— C\J LT)
I I
o o
in in if)
^ co d
C 9 w,
cl ai -
LUXFO<CLQ_Q_<
E
— cJ o c c
a> — t
E ™
3 F
CO
I Ills
ffl ™ (fl
*^ J_l .= O 14 t id
F O SUOidUQ
I Si
o o co _j
Q. CC CQ CL
S I s .s
h- Z Z I- D
(D _J Q- O
_- = >:.
X X X X X
> >
XX —
£ >>>>SSxxxxxx _ = 5>>>>>
712
EXPLANATORY SUPPLEMENT
Table 15.11
planetary Rings
Saturn Ring Data
Feature Distance (km) Distance (Rs)
Equatorial radius
60330
1.000
D-ring inner edge
67000
1.11
C-ring inner edge
74400
1,233
B-ring inner edge
91900
1.524
B-ring outer edge
117400
1.946
A-ring inner edge
121900
2.021
A-ring gap center
133400
2.212
A-ring outer edge
136600
2.265
F-ring center
140300
2.326
G-ring center
170000
2.8
E-ring inner edge
- 180000
~3
E-ring outer edge
- 480000
-8
Rings of Uranus
Ring Semi-major Axis (km) Eccentricity Azimuth of Precession Rate (deg./day)
Periapse (deg.)
6
41870
0.0014
236
2.77
5
42270
0.0018
182
2.66
4
42600
0.0012
120
2.60
a
44750
0.0007
331
2.18
0
45700
0.0005
231
2.03
V
47210
7
47660
s
48330
0.0005
140
51180
0.0079
216
1.36
Epoch: 1977 March 10, 20h UT (JD 2443213.33)
15 / REFERENCE DATA
713
Table 15.12
Constellation Names and Abbreviations
I^Ulllll ItJLI VB
HDurev.
Genitive
Andromeda
And
Andromedae
Antlia
Ant
Antliae
Apus
Aps
Apodis
Aquarius
Aqr
Ao,uarii
Aquiia
Aql
Aqutlae
Ara
Ara
Arae
*Argo
Arg
Argus
Aries
AH
A net is
Auriga
Aur
Ai i rinap
fl U J lUuw
Bootes
Boo
Bo otis
Caelum
Cae
Caeli
Camelopardalis
Cam
Camelopardalis
Cancer
Cnc
Cancri
Canes Venatici
CVn
Canum Venaticorum
Canls Major
CMa
Ganis Majohs
Canis Minor
CMi
Canis Minoris
Capri corn us
Cap
Capri comi
Carina
Car
Cassiopeia
Cas
Cassiope ae
Centaurus
Cen
Centauri
Cepheus
Cep
Cephei
Cetus
Cet
Ceti
Chamaeleon
Cha
Chamaeleontis
Circinus
Cir
i Tr- i n i
W 1 U - 1 II
Columba
Col
Columbae
Coma Berenices
Com
Comae Berenices
t Corona Austrina CrA
Coronae Austrinae
Corona Borealis
CrB
Coronae Borealis
Corvus
Crv
. Corvi
Crater
Crt
C, r 3f Pr i
Crux
Cru
Crucis
Cygnus
Cyg
Cygni
Delphinus
Del
Delphini
Dorado
Dor
Doradus
Draco
Dra
Equuleus
Equ
Equulei
Eridanus
Eri
Eridani
Fornax
For
Fornacis
Gemini
Gem
Geminorum
Gms
Gru
Gruis
Hercules
Her
Hercuiis
Horologium
Hor
Horologii
Hydra
Hya
Hydrae
Hydrus
Hyi
Hydri
Indus
Ind
Indi
Nominative Abbrev. Genitive
Lacerta
Lac
Lacertae
Leo
Leo
Leonis
Leo Minor
LMi
Leonis Minoris
Lepus
Lep
Leporis
Libra
Lib
Librae
Lupus
Lup
Lupi
Lynx
Lyn
Lyncis
Lyra
Lyr
Lyrae
Mensa
Men
Mensae
Microscopium
Mic
Microscopii
Monoceros
Mon
Monocerotis
Musca
Mus
Muscae
Norma
Nor
Normae
Octans
Oct
Octarttis
Oph iuchus
\jpn
Ophiuchi
Orion
Ori
Ononis
Pavo
Pav
Pavonis
Peg
Pegasi
Perseus
Per
Persei
Phoenix
Phe
Phoenicis
Pictor
Pic
Picloris
Pisces
Psc
Piscium
^Piscis Austrinus
PsA
Piscis Austrini
Puppis
Pup
Puppis
Pyxis
Pyx
Pyxidis
Reticulum
Ret
Reticuli
oagrna
Sge
Sagittae
Sagittarius
Sgr
Sagittarii
Scorprus
Sco
Scorpii
Sculptor
Scl
Sculptoris
Scutum
Set
Scuti
^Serpens
Ser
Serpentis
Sextans
Sex
Sextantis
Taurus
Tau
Tauri
Telescopium
Tel
Telescopii
Triangulum
Tn
Trianguli
Triangulum Austrate TrA
Trianguli Australis
Tucana
Tuc
Tucanae
Ursa Major
UMa
Ursae Majoris
Ursa Minor
UMi
Ursae Minoris
Vela
Vel
Velorum
Virgo
Vir
Virginls
Volans
Vol
Volantis
Vulpecula
Vul
Vulpeculae
Note: This list of constellation names and abbreviations is in accordance with the resolutions of the International
Astronomical Union [Trans. IAUA, 156: 4, 221; 9, 66 and 77). The boundaries of the constellations are listed by E.
Delporte, on behalf of the IAU, in Delimitation scientifique des constellations (tables et cartes) (Cambridge
University Press, 1930); the areas of the constellations are given in Handbook B.A.A., 1961.
* In modern usage Argo is divided into Carina, Puppis, and Vela.
* Australis is sometimes used, in both nominative and genitive.
' Serpens may be divided into Serpens Caput and Serpens Cauda.
4
EXPLANATORY SUPPLEMENT
Table 15.13
Mathematical Constants
v =3 141592653589793
e =2.718281828459045
=23.140692632779269
log10 X =0. 4342944819032518 loge X
|0ge x = 2. 302585092994046 log10 X
1 radian = 57?29577951 30823
= 3437'.74677078494
= 206264"806247096
1 ° = 0. 01 745329251 99433 radian
1 ' =0. 000290888208665722
r =0.00000484813681109536
Solid Angles
1 steradian = 3283 square degrees 1 square degree = 0. 305 x 10^3 steradians
1 steradian = 1. 18 x 107 square minules 1 square minute = 0.846 x 10 ~7 steradians
1 steradian = 4.25 x 1010 square seconds 1 square second = 0.235 x 10 steradians
A sphere subtends 4ir steradians = 41253 square degrees
= 1.485 x 10B square minutes = 5.35 x 1011 square seconds
716
EXPLANATORY SUPPLEMENT
It
c. —
S c
»
II
M
ffl
T3
C
to
03
03
Q.
CO
m c
i- a)
ui ->
a
* to
li
1
o
1
o
I
o
X
X
X
0)
^•
CO
o
CM
5
o>
CD
US
r~-
CM
r>
S
lO
o
in
CO
^■
«
s
m"
lo
«
7
T
icy
o
o
I
O
X
X
X
t- t-
i- o ^-
X X
x CO
hi ifi
m
o m t-
co cvj in
S -5
* (I L
8 f g
o
a> e
i _
sis
ill
1 *f
|3 *" "S
< 3 ^,
LT]
O
1
Si
o
I
O
£
i
S. S i=
o
c\i i-
CP
Si!
i- CM t-
r 8 5
V» T- *• > O
to
S r-
£ a)
"I" B
<E0 E
<D CO O CO
= 3 co in
e: o
rt in cm
3 II CD O
2 3 11 ci
§ S t ii
T- T- T- O
9>
15 / REFERENCE DATA
717
Table 15.16
Greek Alphabet
Of
a Inhs
til \J • 1 K±
'I
H
eta
u
N
nu
T
T
tau
0
B
beta
e
theta
%
xi
V
Y
upsilon
7
r
gamma
i
iota
0
O
omicron
4
*
phi
6
A
delta
K
K
kappa
ir
n
pi*
X
X
chi
£
E
epsilon
A
A
lambda
P
p
mo
4
*
psi
c
Z
zeta
M
M
mu
a
E
sigmaf
Q
omega
* m ('curly pi') is an alternative form of tt
t t is an alternative form of a.
718
EXPLANATORY SUPPLEMENT
Table 15.17
International System of Units (SI)
SI Base Units
Quantity
Name
Symbol
length
meter
m
mass
kilogram
kq
time
second
s
electric current
ampere
A
thermodynamic temperature
kelvin
K
amount of substance
mole
mol
luminous intensity
candela
cd
SI Derived Units with Special Names
Quantity
Name
Symbol
Expression
Expression
in terms
in terms
of other units
of SI base units
frequency
hertz
Hz
s~1
force
newton
N
m If n c ^
1 1 1 r\y a
pressure, stress
pascal
Pa
N/m2
energy, work quantity
of heat
joule
N m
power, radiant flux
watt
W
.l/<s
1 I 1 "
electric charge,
quantity of electricity
coulomb
c
s A
electric potential,
potential difference,
electromotive force
volt
v
V
W/A
m2 L-,-1 o — 3 J — 1
ni f\y o r\
capacitance
farad
F
CN
m-2kg-1s4A2
electric resistance
ohm
n
V/A
m-2kgs-3A-2
electric conductance
Siemens
s
A/V
m~2 kg~1s3A2
magnetic flux
weber
Wb
Vs
m2 kg s— 2 A-1
magnetic flux density
tesla
T
Wb/m2
kgs"2A"1
inductance
henry
H
Wb/A
m2 kgs_2A— 2
Celsius temperature
degree Celsius
°C
K
luminous flux
lumen
Im
cd sr
illuminance
lux
Ix
lm/m2
m-2 cd sr
SI Supplementary Units
Quantity
Name
Symbol
plane angle
radian
rad
solid angle
steradian
sr
15 / REFERENCE DATA
719
Table 15.17, continued
International System of Units (SI)
Units in Use with the International System
Name
Symbol
Value in Si unit
minute
min
1 min =
60s
hour
h
1 h
=
60 min =
3600 s
day
d
1d
=
24 h = 86400 s
degree
0
1°
=
<7r/ 180) rad
minute
/
r
_
(1 /60)°
= (tt/ 10800) rad
second
//
1"
=
(1 / 60)'
= (tt / 648000) rad
liter
I, L
1 L
=
1 rim3 -
i urn —
1 U ll\
metric ton t
1 1
103 kg
SI Prefixes
Factor
Prefix
Symbol
Factor
Prefix
Symbol
1018
exa
E
1CT1
deci
d
101S
peta
P
10"2
centi
c
1012
tera
T
10-3
milli
m
109
giga
G
10~6
micro
M
106
mega
M
io-9
nano
n
103
kilo
k
10"12
pico
P
10s
hecto
h
10-15
femto
f
101
deka
da
10-18
atto
a
15.1 REFERENCES
Taylor, B.N. and Cohen, E.R. (1986). "The 1986 Adjustment of the Fundamental Physical
Constants" COD ATA Newsletter, November.
Davies, M.E. et al. (1989). "Report of the IAU/IAG/COSPAR Working Group on Car-
tographic Coordinates and Rotational Elements of the Planets and Satellites: 1988"
Celestial Mechanics 46, 187-204.
Moritz, H. (1984). "Geodetic Reference System 1980" Bull GeodesiqueSS, 388-398.
(1986). The International System of Units (SI). Goodman, D.T. and Bell, R.J.,
eds. U.S. Department of Commerce.
Glossary
aberration the apparent angular displacement of the observed position of a celestial
object; from its geometric position, caused by the finite velocity of light in
combination with the motions of the observer and of the observed object.
(See aberration, planetary.)
aberration, annual the component of stellar aberration (see aberration, stellar) re-
sulting from the motion of the Earth about the Sun.
aberration, diurnal the component of stellar aberration (see aberration, stellar)
resulting from the observer's diurnal motion about the center of the Earth.
aberration, E-terms of terms of annual aberration (see aberration, annual) depend-
ing on the eccentricity and longitude of perihelion (see longitude; pericenter)
of the Earth.
aberration, elliptic see aberration, E-terms of.
aberration, planetary the apparent angular displacement of the observed position
of a celest ial body produced by motion of the observer (see aberration, stellar)
and the actual motion of the observed object (see correction for light-time).
aberration, secular the component of stellar aberration (see aberration, stellar)
resulting from the essentially uniform and rectilinear motion of the entire
solar system in space. Secular aberration is usually disregarded.
aberration, stellar the apparent angular displacement of the observed position of
a celestial body resulting from the motion of the observer. Stellar aberra-
tion is divided into diurnal, annual, and secular components. (See aberration,
diurnal; aberration, annual; aberration, secular.)
721
y22 GLOSSARY
altitude the angular distance of a celestial body above or below the horizon, mea-
sured along the great circle passing through the body and the zenith. Altitude
is 90° minus zenith distance.
anomaly angular measurement of a body in its orbit from its perihelion.
aphelion the point in a planetary orbit that is at the greatest distance from the
Sun.
apogee the point at which a body in orbit around the Earth reaches its farthest
distance from the Earth.
apparent place the position on a celestial sphere, centered at the Earth, determined
by removing from the directly observed position of a celestial body the effects
that depend on the topocentric location of the observer; i.e., refraction, di-
urnal aberration (see aberration, diurnal), and geocentric (diurnal) parallax.
Thus the position at which the object would actually be seen from the center
of the Earth, displaced by planetary aberration (except the diurnal part — see
aberration, planetary; aberration, diurnal) and referred to the true equator
and equinox.
apparent solar time the measure of time based on the diurnal motion of the true
Sun. The rate of diurnal motion undergoes seasonal variation because of the
obliquity of the ecliptic and because of the eccentricity of the Earth's orbit.
Additional small variations result from irregularities in the rotation of the
Earth on its axis.
aspect the apparent position of any of the planets or the Moon relative to the Sun,
as seen from Earth.
astrometric ephemeris an ephemeris of a solar-system body in which the tabulated
positions are essentially comparable to catalog mean places of st ars at a stan-
dard epoch. An astrometric position is obtained by adding to the geometric
position, computed from gravitational theory, the correction for light-time.
Prior to 1984, the E-terms of annual aberration (see aberration, annual: aber-
ration, E-terms of) were also added to the geometric position.
astronomical coordinates the longitude and latitude of a point on t he Earth relative
to the geoid. These coordinates are influenced by local gravity anomalies. (See
zenith; longitude, terrestrial; latitude, terrestrial.)
astronomical unit (au) the radius of a circular orbit in which a body of negligible
mass, and free of perturbations, would revolve around the Sun in 2n I k days,
where k is the Gaussian gravitational constant. This is slightly less than the
semi-major axis of the Earth's orbit.
GLOSSARY
atomic second see second, Systeme International.
723
augmentation the amount by which the apparent semidiameter of a celestial body
as observed from the surface of the Earth, is greater than the semidiameter
that would be observed from the center of the Earth.
azimuth the angular distance measured clockwise along the horizon from a specified
reference point (usually north) to the intersection with the great circle drawn
from the zenith through a body on the celestial sphere.
barycenter the center of mass of a system of bodies; e.g., the center of mass of the
solar system or the Earth-Moon system.
Barycentric Dynamical Time (TDB) the independent argument of ephemerides and
equations of motion that are referred to the barycenter of the solar system. A
family of timescales results from the transformation by various theories and
metrics of relativistic theories of Terrestrial Dynamical Time (TDT). TDB
differs from TDT only by periodic variations. In the terminology of the gen-
eral theory of relativity, TDB may be considered to be a coordinate time. (See
dynamical time.)
brilliancy for Mercury and Venus the quantity ks2 I r2, where k = 0.5(1 + cos/), i
is the phase angle, s is the apparent semidiameter, and r is the heliocentric
distance.
calendar a system of reckoning time in which days are enumerated according to
their position in cyclic patterns.
catalog equinox the intersection of the hour circle of zero right ascension of a star
catalog with the celestial equator. (See dynamical equinox; equator.)
celestial ephemeris pole the reference pole for nutation and polar motion: the axis
of figure for the mean surface of a model Earth in which the free motion has
zero amplitude. This pole has no nearly diurnal nutation with respect to a
space- fixed or Earth-fixed coordinate system.
celestial equator the projection onto the celestial sphere of the Earth's equator.
(See mean equator and equinox; equinox; true equator and equinox.)
celestial pole either of the two points projected onto the celestial sphere by the
extension of the Earth's axis of rotation to infinity.
celestial sphere an imaginary sphere of arbitrary radius upon which celestial bodies
may be considered to be located. As circumstances require, the celestial sphere
may be centered at the observer, at the Earth's center, or at any other location.
724 GLOSSARY
conjunction the phenomenon in which two bodies have the same apparent celes-
tial longitude (see longitude, celestial) or right ascension as viewed from a
third body. Conjunctions are usually tabulated as geocentric phenomena. For
Mercury and Venus, geocentric inferior conjunction occurs when the planet is
between the Earth and Sun, and superior conjunction occurs when the Sun is
between the planet and Earth.
constellation a grouping of stars, usually with pictorial or mythical associations,
that serves to identify an area of the celestial sphere. Also, one of the precisely
defined areas of the celestial sphere, associated with a grouping of stars, that
the International Astronomical Union has designated as a constellation.
Coordinated Universal Time (UTC) the timescale available from broadcast time
signals. UTC differs from TAI (see International Atomic Time) by an integral
number of seconds; it is maintained within +0.90 second of UT1 (see Univer-
sal Time) by the introduction of one second steps (leap seconds). (See leap
second.)
culmination passage of a celestial object across the observer's meridian; also called
"meridian passage." More precisely, culmination is the passage through the
point of greatest altitude in the diurnal path. Upper culmination (also called
''culmination above pole" for circumpolar stars and the Moon) or transit is
the crossing closer to the observer's zenith. Lower culmination (also called
"culmination below pole" for circumpolar stars and the Moon) is the crossing
farther from the zenith.
day an interval of 86400 SI seconds (see second, Systfeme International), unless
otherwise indicated.
day numbers quantities that facilitate hand calculations of the reduction of mean
place to apparent place. Besselian day numbers depend solely on the Earth's
position and motion; second-order day numbers, used in higher precision re-
ductions, depend on the positions of both the Earth and the star.
declination angular distance on the celestial sphere north or south of the celestial
equator. It is measured along the hour circle passing through the celestial
object. Declination is usually given in combination with right ascension or
hour angle.
defect of illumination the angular amount of the observed lunar or planetary disk
that is not illuminated to an observer on the Earth.
deflection of light the angle by which I he apparent path of a photon is altered
from a st raight line by the gravitational field of the Sun. The path is deflected
radially away from the Sun by up to 1775 at the Sun's limb. Correction for
GLOSSARY
725
this effect, which is independent of wavelength, is included in the reduction
from mean place to apparent place.
deflection of the vertical the angle between the astronomical vertical and the geode-
tic vertical, (See zenith; astronomical coordinates; geodetic coordinates.)
Delta T (AT) the difference between dynamical time and Universal Time; specif-
ically the difference between Terrestrial Dynamical Time (TDT) and UTl-
A7"=TDT-UT1.
direct motion for orbital motion in the solar system, motion that is counterclock-
wise in the orbit as seen from the north pole of the ecliptic; for an object
observed on the celestial sphere, motion that is from west to east, resulting
from the relative motion of the object and the Earth.
diurnal motion the apparent daily motion of celestial bodies across the sky from
east to west, caused by the Earth's rotation.
AUT1 the predicted value of the difference between UT1 and UTC, transmitted in
code on broadcast time signals: AUT1 = UT1 - UTC. (See Universal Time;
Coordinated Universal Time.)
dynamical equinox the ascending node of the Eartlvs mean orbit on the Earth's
equator: i.e., the intersection of the ecliptic with the celestial equator at which
the Sun's declination is changing from south to north. (See catalog equinox;
equinox.)
dynamical time the family of timescales introduced in 1984 to replace ephemeris
time as the independent argument of dynamical theories and ephemerides.
(See Bary centric Dynamical Time; Terrestrial Dynamical Time )
eccentric anomaly in undisturbed elliptic motion, the angle measured at the center
of the ellipse from pericenter to the point on the circumscribing auxiliary circle
from which a perpendicular to the major axis would intersect the orbiting
body. (See mean anomaly; true anomaly.)
eccentricity a parameter that specifies the shape of a conic section; one of the
standard elements used to describe an elliptic orbit. (See elements, orbital.)
eclipse the obscuration of a celestial body caused by its passage through the shadow
cast by another body.
eclipse, annular a solar eclipse (see eclipse, solar) in which the solar disk is never
completely covered but is seen as an annulus or ring at maximum eclipse. An
annular eclipse occurs when the apparent disk of the Moon is smaller than
that of the Sun.
726 GLOSSARY
eclipse, lunar an eclipse in which the Moon passes through the shadow cast by
the Earth. The eclipse may be total (the Moon passing completely through
the Earth's umbra), partial (the Moon passing partially through the Earth's
umbra at maximum eclipse), or penumbral (the Moon passing only through
the Earth's penumbra).
eclipse, solar an eclipse in which the Earth passes through the shadow cast by the
Moon. It may be total (observer in the Moon's umbra), partial (observer in
the Moon's penumbra), or annular. (See eclipse, annular.)
ecliptic the mean plane of the Earth's orbit around the Sun.
elements, Besselian quantities tabulated for the calculation of accurate predictions
of an eclipse or occultation for any point on or above the surface of the Earth.
elements, orbital parameters that specify the position and motion of a body in
orbit. (See osculating elements; mean elements.)
elongation, greatest the instants when the geocentric angular distances of Mercury
and Venus are at a maximum from the Sun.
elongation (planetary) the geocentric angle between a planet and the Sun, mea-
sured in the plane of the planet, Earth and Sun. Planetary elongations are
measured from 0° to 180°, east or west of the Sun.
elongation (satellite) the geocentric angle between a satellite and its primary, mea-
sured in the plane of the satellite, planet and Earth. Satellite elongations are
measured from 0° east or west of the planet.
epact the age of the Moon; the number of days since New Moon, diminished by one
day, on January 1 in the Gregorian ecclesiastical lunar cycle. (See Gregorian
calendar and lunar phases.)
ephemeris a tabulation of the positions of a celestial object in an orderly sequence
for a number of dates.
ephemeris hour angle an hour angle referred to the ephemeris meridian.
ephemeris longitude longitude (see longitude, terrestrial) measured eastward from
the ephemeris meridian.
ephemeris meridian a fictitious meridian that rotates independently of the Earth
at the uniform rate implicitly defined by Terrestrial Dynamical Time (TDT).
The ephemeris meridian is 1. 002738/17" east of the Greenwich meridian, where
AT = TDT — UT1.
GLOSSARY 727
ephemeris time (ET) the timescale used prior to 1984 as the independent, variable
in gravitational theories of the solar system. In 1984, ET was replaced by
dynamical time.
ephemeris transit the passage of a celestial body or point across the ephemeris
meridian.
epoch an arbitrary fixed instant of time or date used as a chronological reference
datum for calendars (see calendar), celestial reference systems, star catalogs,
or orbital motions (see orbit).
equation of center in elliptic motion the true anomaly minus the mean anomaly. It
is the difference between the actual angular position in the elliptic orbit and
the position the body would have if its angular motion were uniform.
equation of the equinoxes the right ascension of the mean equinox (see mean equa-
tor and equinox) referred to the true equator and equinox; apparent sidereal
time minus mean sidereal time. (See apparent place; mean place.)
equation of time the hour angle of the true Sun minus the hour angle of the ficti-
tious mean sun; alternatively, apparent solar time minus mean solar time.
equator the great circle on the surface of a body formed by the intersection of the
surface with the plane passing through the center of the body perpendicular
to the axis of rotation. (See celestial equator.)
equinox either of the two points on the celestial sphere at which the ecliptic in-
tersects the celestial equator; also the time at which the Sun passes through
either of these intersection points; ie., when the apparent longitude (see appar-
ent place; longitude, celestial) of the Sun is 0° or 180°. (See catalog equinox;
dynamical equinox for precise usage.)
era a system of chronological notation reckoned from a given date.
fictitious mean sun an imaginary body introduced to define mean solar time; es-
sentially the name of a mathematical formula that defined mean solar time.
This concept is no longer used in high precision work.
flattening a parameter that specifies the degree by which a planet's figure differs
from that of a sphere; the ratio / = (a — b) la, where a is the equatorial radius
and b is the polar radius.
frequency the number of cycles or complete alternations per unit time of a carrier
wave, band, or oscillation.
728 GLOSSARY
frequency standard a generator whose output is used as a precise frequency refer-
ence; a primary frequency standard is one whose frequency corresponds to the
adopted definition of the second (see second, Systeme International), with its
specified accuracy achieved without calibration of the device.
Gaussian gravitational constant (k = 0.01720209895) the constant defining the as-
tronomical system of units of length (astronomical unit), mass (solar mass)
and time (day), by means of Kepler's third law. The dimensions of k2 are
those of Newton's constant of gravitation:
gegenschein faint nebulous light about 20° across near the ecliptic and opposite
the Sun, best seen in September and October. Also called counterglow.
geocentric with reference to, or pertaining to, the center of the Earth.
geocentric coordinates the latitude and longitude of a point on the Earth's surface
relative to the center of the Earth; also celestial coordinates given with re-
spect to the center of the Earth. (See zenith; latitude, terrestrial; longitude,
terrestrial.)
geodetic coordinates the latitude and longitude of a point on the Earth's surface
determined from the geodetic vertical {normal to the specified spheroid). (See
zenith; latitude, terrestrial; longitude, terrestrial.)
geoid an equipotential surface that coincides with mean sea level in the open ocean.
On land it is the level surface that would be assumed by water in an imaginary
network of frictionless channels connected to the ocean.
geometric position the geocentric position of an object on the celestial sphere re-
ferred to the true equator and equinox, but without the displacement due to
planetary aberration. (See apparent place; mean place; aberration, planetary.)
Greenwich sidereal date (GSD) the number of sidereal days elapsed at Greenwich
since the beginning of the Greenwich sidereal day that was in progress at
Julian date 0.0.
Greenwich sidereal day number the integral part of the Greenwich sidereal date.
Gregorian calendar the calendar introduced by Pope Gregory XIII in 1582 to re-
place the Julian calendar; the calendar now used as the civil calendar in most
countries. Every year that is exactly divisible by four is a leap year, except
for centurial years, which must be exactly divisible by 400 to be leap years.
Thus 2000 is a leap year, but 1900 and 2100 are not leap years.
height elevation above ground or distance upwards from a given level (especially
sea level) to a fixed point. (See altitude.)
GLOSSARY 729
heliocentric with reference to, or pertaining to, the center of the Sun.
horizon a plane perpendicular to the line from an observer to the zenith. The great
circle formed by the intersection of the celestial sphere with a plane perpen-
dicular to the line from an observer to the zenith is called the astronomical
horizon.
horizontal parallax the difference between the topocentric and geocentric positions
of an object, when the object is on the astronomical horizon.
hour angle angular distance on the celestial sphere measured westward along the
celestial equator from the meridian to the hour circle that passes through a
celestial object.
hour circle a great circle on the celestial sphere that passes through the celestial
poles and is therefore perpendicular to the celestial equator.
inclination the angle between two planes or their poles; usually the angle between
an orbital plane and a reference plane; one of the standard orbital elements
(see elements, orbital) that specifies the orientation of an orbit.
International Atomic Time (TAI) the continuous scale resulting from analyses by
the Bureau International des Poids et Mesures of atomic time standards in
many countries. The fundamental unit of TAI is the SI second (see second,
Systeme International), and the epoch is 1958 January 1.
invariable plane the plane through the center of mass of the solar system perpen-
dicular to the angular momentum vector of the solar system.
irradiation an optical effect of contrast that makes bright objects viewed against a
dark background appear to be larger than they really are.
Julian calendar the calendar introduced by Julius Caesar in 46 B.C. to replace the
Roman calendar. In the Julian calendar a common year is defined to comprise
365 days, and every fourth year is a leap year comprising 366 days. The Julian
calendar was superseded by the Gregorian calendar.
Julian date (JD) the interval of time in days and fraction of a day since 4713 B.C.
January 1, Greenwich noon, Julian proleptic calendar. In precise work the
timescale, e.g., dynamical time or Universal Time, should be specified.
Julian date, modified (MJD) the Julian date minus 2400000.5.
Julian day number (JD) the integral part of the Julian date.
Julian proleptic calendar the calendric system employing the rules of the Julian
calendar, but extended and applied to dates preceding the introduction of the
Julian calendar.
730 GLOSSARY
Julian year a period of 365.25 days. This period served as the basis for the Julian
calendar.
Laplacian plane for planets see invariable plane; for a system of satellites, the fixed
plane relative to which the vector sum of the disturbing forces has no orthog-
onal component.
latitude, celestial angular distance on the celestial sphere measured north or south
of the ecliptic along the great circle passing through the poles of the ecliptic
and the celestial object.
latitude, terrestrial angular distance on the Earth measured north or south of the
equator along the meridian of a geographic location.
leap second a second (see second, Systeme International) added between 60s and 0s
at announced times to keep UTC within 0*90 of UT1. Generally, leap seconds
are added at the end of June or December.
librations variations in the orientation of the Moon's surface with respect to an
observer on the Earth. Physical librations are due to variations in the rate
at which the Moon rotates on its axis. The much larger optical librations are
due to variations in the rate of the Moon's orbital motion, the obliquity of
the Moon's equator to its orbital plane, and the diurnal changes of geometric
perspective of an observer on the Earth's surface.
light, deflection of the bending of the beam of light due to gravity. It is observable
when the light from a star or planet passes a massive object such as the Sun.
light-time the interval of time required for light to travel from a celestial body
to the Earth. During this interval the motion of the body in space causes
an angular displacement of its apparent place from its geometric place (see
geometric position). (See aberration, planetary.)
light-year the distance that light, traverses in a vacuum during one year.
limb the apparent edge of the Sun, Moon, or a planet or any other celestial body
with a detectable disc.
limb correction correction that must be made to the distance between the center
of mass of the Moon and its limb. These corrections are due to the irregular
surface of the Moon and are a function of the librations in longitude (see
longitude, celestial) and latitude (see latitude, celestial) and the position angle
from the central meridian.
local sidereal time the local hour angle of a catalog equinox.
GLOSSARY
731
longitude, celestial angular distance on the celestial sphere measured eastward
along the ecliptic from the dynamical equinox to the great circle passing
through the poles of the ecliptic and the celestial object.
longitude, terrestrial angular distance measured along the Earth's equator from
the Greenwich meridian to the meridian of a geographic location.
luminosity class distinctions among stars of the same spectral class. (See spectral
types or classes.)
lunar phases cyclically recurring apparent forms of the Moon. New Moon, First
Quarter, Full Moon, and Last Quarter are defined as the times at which the
excess of the apparent celestial longitude (see longitude, celestial) of the Moon
over that of the Sun is 0°, 90°, 180°, and 270°, respectively.
lunation the period of time between two consecutive New Moons.
magnitude, stellar a measure on a logarithmic scale of the brightness of a celestial
object considered as a point source.
magnitude of a lunar eclipse the fraction of the lunar diameter obscured by the
shadow of the Earth at the greatest phase of a lunar eclipse (see eclipse,
lunar), measured along the common diameter.
magnitude of a solar eclipse the fraction of the solar diameter obscured by the
Moon at the greatest phase of a solar eclipse (see eclipse, solar), measured
along the common diameter.
mean anomaly in undisturbed elliptic motion, the product of the mean motion of
an orbiting body and the interval of time since the body passed pericenter.
Thus the mean anomaly is the angle from pericenter of a hypothetical body
moving with a constant angular speed that is equal to the mean motion. (See
true anomaly; eccentric anomaly.)
mean distance the semi- major axis of an elliptic orbit.
mean elements elements of an adopted reference orbit (see elements, orbital) that
approximates the actual, perturbed orbit. Mean elements may serve as the
basis for calculating perturbations.
mean equator and equinox the celestial reference system determined by ignoring
small variations of short period in the motions of the celestial equator. Thus
the mean equator and equinox are affected only by precession. Positions in
star catalogs are normally referred to the mean catalog equator and equinox
(see catalog equinox) of a standard epoch.
732 GLOSSARY
mean motion in undisturbed elliptic motion, the constant angular speed required
for a body to complete one revolution in an orbit of a specified semi-major
axis.
mean place the coordinates, referred to the mean equator and equinox of a stan-
dard epoch, of an object on the celestial sphere centered at the Sun. A mean
place is determined by removing from the directly observed position the ef-
fects of refraction, geocentric and stellar parallax, and stellar aberration (see
aberration, stellar), and by referring the coordinates to the mean equator
and equinox of a standard epoch. In compiling star catalogs it has been the
practice not to remove the secular part of stellar aberration (see aberration,
secular). Prior to 1984, it was additionally the practice not to remove the el-
liptic part of annual aberration (see aberration, annual; aberration, E-terms
of).
mean solar time a measure of time based conceptually on the diurnal motion of t he
fictitious mean sun, under the assumption that the Earth's rate of rotation is
constant.
meridian a great circle passing through the celestial poles and through the zenith
of any location on Earth. For planetary observations a meridian is half the
great circle passing through the planet's poles and through any location on
the planet.
month the period of one complete synodic or sidereal revolution of the Moon around
the Earth; also a calendrical unit that approximates the period of revolution.
moonrise, moonset the times at which the apparent upper limb of the Moon is on
the astronomical horizon; i.e., when the true zenith distance, referred to the
center of the Earth, of the central point of the disk is 90°34' +s - jr, where s is
the Moon's semidiameter, 7r is the horizontal parallax, and 34' is the adopted
value of horizontal refraction.
nadir the point on the celestial sphere diametrically opposite to the zenith.
node either of the points on the celestial sphere at which the plane of an orbit
intersects a reference plane. The position of a node is one of the standard
orbital elements (see elements, orbital) used to specify the orientation of an
orbit.
nutation the short-period oscillations in the motion of the pole of rotation of a
freely rotating body that is undergoing torque from external gravitational
forces. Nutation of the Earth's pole is discussed in terms of components in
obliquity and longitude (see longitude, celestial.)
GLOSSARY 733
obliquity in general the angle between the equatorial and orbital planes of a body
or, equivalently, between the rotational and orbital poles. For the Earth the
obliquity of the ecliptic is the angle between the planes of the equator and
the ecliptic.
occultation the obscuration of one celestial body by another of greater apparent
diameter; especially the passage of the Moon in front of a star or planet, or
the disappearance of a satellite behind the disk of its primary. If the primary
source of illumination of a reflecting body is cut off by the occultation, the
phenomenon is also called an eclipse. The occultation of the Sun hv the Moon
is a solar eclipse (see eclipse, solar. )
opposition a configuration of the Sun, Earth and a planet in which the apparent
geocentric longitude (see longitude, celestial) of the planet differs by 180°
from the apparent geocentric longitude of the Sun.
orbit the path in space followed by a celestial body.
osculating elements a set of parameters (see elements, orbital) that specifies the
instantaneous position and velocity of a celestial body in its perturbed orbit.
Osculating elements describe the unperturbed (two-body) orbit that the body
would follow if perturbations were to cease instantaneously.
parallax the difference in apparent direction of an object as seen from two different
locations; conversely, the angle at the object, that is subtended by the line
joining two designated points. Geocentric (diurnal) parallax is the difference
in direction between a topocentric observation and a hypothetical geocentric
observation. Heliocentric or annual parallax is the difference between hypo-
thetical geocentric and heliocentric observations; it is the angle subtended at
the observed object by the semi-major axis of the Earth's orbit. (See also
horizontal parallax.)
parsec the distance at which one astronomical unit subtends an angle of one second
of arc; equivalently. the distance to an object having an annual parallax of
one second of arc.
penumbra the portion of a shadow in which light from an extended source is par-
tially but not completely cut off by an intervening body; the area of partial
shadow surrounding the umbra.
pericenter the point in an orbit that is nearest to the center of force. (See perigee;
perihelion.)
perigee the point, at. which a body in orbit around the Earth most closely ap-
proaches the Earth. Perigee is sometimes used with reference to the apparent
orbit of the Sun around the Earth.
734 GLOSSARY
perihelion the point at which a body in orbit around the Sun most closely ap-
proaches the Sun.
period the interval of time required to complete one revolution in an orbit or one
cycle of a periodic phenomenon, such as a cycle of phases. (See phase.)
perturbations deviations between the actual orbit of a celestial body and an as-
sumed reference orbit; also, the forces that cause deviations between the actual
and reference orbits. Perturbations, according to the first meaning, are usually
calculated as quantities to be added to the coordinates of the reference orbit
to obtain the precise coordinates.
phase the ratio of the illuminated area of the apparent disk of a celestial body to
the area of the entire apparent disk taken as a circle. For the Moon, phase
designations (see lunar phases) are defined by specific configurations of the
Sun, Earth and Moon. For eclipses, phase designations (total, partial, penum-
bral, etc.) provide general descriptions of the phenomena. (See eclipse, solar:
eclipse, annular; eclipse, lunar.)
phase angle the angle measured at the center of an illuminated body between the
light source and the observer.
photometry a measurement of the intensity of light usually specified for a specific
frequency range.
planetocentric coordinates coordinates for general use, where the z-axis is the mean
axis of rotation; the .r-axis is the intersection of the planetary equator (normal
to the --axis through the center of mass) and an arbitrary prime meridian: and
they-axis completes a right-hand coordinate system. Longitude (see longitude,
celestial) of a point is measured positive to the prime meridian as defined by
rotational elements. Latitude (see latitude, celestial) of a point is the angle
between the planetary equator and a line to the center of mass. The radius is
measured from the center of mass to the surface point.
planetographic coordinates coordinates for cartographic purposes dependent on an
equipotential surface as a reference surface. Longitude (see longitude, celes-
tial) of a point is measured in the direction opposite to the rotation (positive
to the west for direct rotation) from the cartographic position of the prime
meridian defined by a clearly observable surface feature. Latitude (see lati-
tude, celestial) of a point is the angle between the planetary equator (normal
to the z-axis and through the center of mass) and normal to the reference
surface at the point. The height of a point is specified as the distance above
a point with the same longitude and latitude on the reference surface.
polar motion the irregularly varying motion of the Earth's pole of rotation with
respect to the Earth's crust. (See celestial ephemeris pole.)
GLOSSARY 735
precession the uniformly progressing motion of the pole of rotation of a freely
rotating body undergoing torque from external gravitational forces. In the case
of the Earth, the component of precession caused by the Sun and Moon acting
on the Earth's equatorial bulge is called lunisolar precession; the component
caused by the action of the planets is called planetary precession. The sum of
lunisolar and planetary precession is called general precession. (See nutation.)
proper motion the projection onto the celestial sphere of the space motion of a
star relative to the solar system; thus the transverse component of the space
motion of a star with respect to the solar system. Proper motion is usually-
tabulated in star catalogs as changes in right ascension and declination per
year or century.
quadrature a configuration in which two celestial bodies have apparent longitudes
(see longitude, celestial) that differ by 90° as viewed from a third body.
Quadratures are usually tabulated with respect to the Sun as viewed from
the center of the Earth.
radial velocity the rate of change of the distance to an object.
refraction, astronomical the change in direction of travel (bending) of a light ray as
it passes obliquely through the atmosphere. As a result of refraction, the ob-
served altitude of a celestial object is greater than its geometric altitude. The
amount of refraction depends on the altitude of the object and on atmospheric
conditions.
retrograde motion for orbital motion in the solar system, motion that is clockwise
in the orbit as seen from the north pole of the ecliptic; for an object observed
on the celestial sphere, motion that is from east to west, resulting from the
relative motion of the object and the Earth. (See direct motion.)
right ascension angular distance on the celestial sphere measured eastward along
the celestial equator from the equinox to the hour circle passing through the
celestial object. Right ascension is usually given in combination with declina-
tion.
Saros a Babylonian lunar cycle of 6585.32 days, or 18 years 11.33 days, or 223
lunations, at the end of which the centers of the Sun and Moon return so
nearly to the relative positions of the beginning that all the eclipses (see
eclipse) of the period recur approximately as before, but in longitudes (see
longitude, terrestrial) approximately 120° to the west. (See lunar phases.)
satellite natural body revolving around a planet.
satellite, artificial device launched into a closed orbit around the Earth, another
planet, the Sun, etc.
736 GLOSSARY
second, Systeme International (SI) the duration of 9192631770 cycles of radiation
corresponding to the transition between two liyperfine levels of the ground
state of cesium 133.
selenocentric with reference to, or pertaining to, the center of the Moon,
semidiameter the angle at the observer subtended by the equatorial radius of the
Sun, Moon, or a planet.
semi-major axis half the length of the major axis of an ellipse; a standard element
used to describe an elliptical orbit (see elements, orbital).
sidereal day the interval of time between two consecutive transits of the catalog
equinox. (See sidereal time.)
sidereal hour angle angular distance on the celestial sphere measured westward
along the celestial equator from the catalog equinox to the hour circle pass-
ing through the celestial object. It is equal to 360° minus right ascension in
degrees.
sidereal time the measure of time defined by the apparent diurnal motion of the
catalog equinox; hence a measure of the rotation of the Earth with respect to
the stars rather than the Sun.
solstice either of the two points on the ecliptic at which the apparent longitude
(see longitude, celestial) of the Sun is 90° or 270°; also the time at which the
Sun is at either point.
spectral types or classes catagorization of stars according to their spectra, primar-
ily due to differing temperatures of the stellar atmosphere. From hottest to
coolest, the spectral types are O, B, A, F, G, K, and M.
standard epoch a date and time that specifies the reference system to which celes-
tial coordinates are referred. Prior to 1984 coordinates of star catalogs were
commonly referred to the mean equator and equinox of the beginning of a
Besselian year (see year, Besselian). Beginning with 1984 the Julian year has
been used, as denoted by the prefix J, e.g., J2000.0.
stationary point (of a planet) the position at which the rate of change of the appar-
ent right ascension (see apparent place) of a planet is momentarily zero.
sunrise, sunset the times at which the apparent upper limb of the Sun is on the
astronomical horizon; i.e., when the true zenith distance, referred to the center
of the Earth, of the central point of the disk is 90° 50', based on adopted values
of 34' for horizontal refraction and 16' for the Sun's semidiameter.
GLOSSARY
737
surface brightness (of a planet) the visual magnitude of an average square arc-second
area of the illuminated portion of the apparent disk.
synodic period for planets, the mean interval of time between successive conjunc-
tions of a pair of planets, as observed from the Sun; for satellites, the mean
interval between successive conjunctions of a satellite with the Sun, as ob-
served from the satellite's primary.
synodic time pertaining to successive conjunctions; successive returns of a planet
to the same aspect as determined by Earth.
Terrestrial Dynamical Time (TDT) the independent argument for apparent geo-
centric ephemerides. At 1977 January ld00h00m00s TAI, the value of TDT was
exactly 1977 January ld0003725. The unit of TDT is 86400 SI seconds at
mean sea level. For practical purposes TDT - TAI + 32?184. (See Barycentric
Dynamical Time; dynamical time; International Atomic Time.)
terminator the boundary between the illuminated and dark areas of the apparent
disk of the Moon, a planet or a planetary satellite.
topocentric with reference to, or pertaining to, a point on the surface of the Earth,
usually with reference to a coordinate system.
transit the passage of a celestial object across a meridian; also the passage of one
celestial body in front of another of greater apparent diameter (e.g., the pas-
sage of Mercury or Venus across the Sun or Jupiter's satellites across its disk);
however, the passage of the Moon in front of the larger apparent Sun is called
an annular eclipse (see eclipse, annular). The passage of a body's shadow
across another body is called a shadow transit; however, the passage of the
Moon's shadow across the Earth is called a solar eclipse. (See eclipse, solar.)
true anomaly the angle, measured at the focus nearest the pericenter of an elliptical
orbit, between the pericenter and the radius vector from the focus to the
orbiting body; one of the standard orbital elements (see elements, orbital).
(See also eccentric anomaly; mean anomaly.)
true equator and equinox the celestial coordinate system determined by the in-
stantaneous positions of the celestial equator and ecliptic. The motion of this
system is due to the progressive effect of precession and the short-term, peri-
odic variations of nutation. (See mean equator and equinox.)
twilight the interval of time preceding sunrise and following sunset (see sunrise,
sunset) during which the sky is partially illuminated. Civil twilight comprises
the interval when the zenith distance, referred to the center of the Earth, of
the central point of the Sun's disk is between 90° 50" and 96°, nautical twilight
738 GLOSSARY
comprises the interval from 96° to 102°, astronomical twilight comprises the
interval from 102° to 108°.
umbra the portion of a shadow cone in which none of the light from an extended
light source (ignoring refraction) can be observed.
Universal Time (UT) a measure of time that conforms, within a close approxima-
tion, to the mean diurnal motion of the Sun and serves as the basis of all civil
timekeeping. UT is formally defined by a mathematical formula as a func-
tion of sidereal time. Thus UT is determined from observations of the diurnal
motions of the stars. The timescale determined directly from such observa-
tions is designated UTO; it is slightly dependent on the place of observation.
When UTO is corrected for the shift in longitude (see longitude, terrestrial) of
the observing station caused by polar motion, the timescale UTl is obtained.
Whenever the designation UT is used in this volume, UTl is implied.
vernal equinox the ascending node of the ecliptic on the celestial equator; also the
time at. which the apparent longitude (see apparent place: longitude, celestial)
of the Sun is 0°. (See equinox.)
vertical apparent direction of gravity at the point of observation (normal to the
plane of a free level surface.)
week an arbitrary period of days, usually seven days; approximately equal to the
number of days counted between the four phases of the Moon. (See lunar
phases.)
year a period of time based on the revolution of the Earth around the Sun. The
calendar year (see Gregorian calendar) is an approximation to the tropical
year (see year, tropical). The anomalistic year is the mean interval between
successive passages of the Earth through perihelion. The sidereal year is the
mean period of revolution with respect to the background stars. (See Julian
year; year, Besselian.)
year, Besselian the period of one complete revolution in right ascension of the
fictitious mean sun, as defined by Newcomb. The beginning of a Besselian
year, traditionally used as as standard epoch, is denoted by the suffix ".0".
Since 1984 standard epochs have been defined by the Julian year rather that
the Besselian year. For distinction, the beginning of the Besselian year is now
identified by the prefix B (e.g., B1950.0).
year, tropical the period of one complete revolution of the mean longitude of the
sun with respect to the dynamical equinox. The tropical year is longer than
the Besselian year (see year, Besselian) by 0?148 7*, where T is centuries from
B1900.0.
GLOSSARY ygg
zenith in general, the point directly overhead on the celestial sphere. The astro-
nomical zenith is the extension to infinity of a plumb line. The geocentric
zenith is defined by the line from the center of the Earth through the observer.
The geodetic zenith is the normal to the geodetic ellipsoid at the observer's
location. {See deflection of the vertical.)
zenith distance angular distance on the celestial sphere measured along the great
circle from the zenith to the celestial object. Zenith distance is 90° minus
altitude.
zodiacal light a nebulous light seen in the east before twilight and in the west
after twilight. It is triangular in shape along the ecliptic with the base on the
horizon and its apex at varying altitudes. It is best seen in middle latitudes
(see latitude, terrestrial) on spring evenings and autumn mornings.
Index
This index does not include entries from Chapter 13 (Historical Information), Chapter 14
(Related Publications), or the Glossary. Material in Chapters 13 and 14 can be located
more effectively through the Table of Contents. The Glossary provides definitions of
nomenclature listed in alphabetical order.
A
Aberration, 23, 127-135, 150
annual, 23, 129-130
constant of, 131, 170
correction for light- time, 23, 393
day numbers, 130, 169
differential, 134
diurnal, 129, 132, 160
elliptic (see E-terms)
E-terms, 131, 169
planetary, 23, 127, 133
secular, 129
stellar, 23, 127-128, 130
Adrastea (Jupiter XV), 409, 708, 710
Airglow, 493
Albedo
geometric of asteroids, 415
Allan variance, 59
Altitude, 14
Altitude and azimuth formulas, 208
Amalthea (Jupiter V), 409, 708, 710
orbital elements, 349
American Association of Variable Star
Observers (AAVSO), 513-514
Analemmic curve, 485
Analytical theories, 27
Ananke (Jupiter XII), 708, 710
state vector, 353
Angular velocity. Earth, 51
Apparent ephemerides, 307-308
Apparent places of fundamental stars, 510
Apparent position, 23, 145-160, 165
Apparent sidereal time, 444
Apparent solar time, 74
Argument of pericenter, 27
Ariel (Uranus I), 709, 711
apparent distance and position angle orbital
elements, 368
physical and photometric data, 412
Aries, first point of (see Equinox)
Ascending node, major planets, 315-316, 704
Asteroids (see Minor planets), 310
Astrographic place, 167
Astrometric position, 23, 166, 308, 31 1-313
Astronomical constants, 280, 289, 302-303,
317-319
coordinates, 16, 202-207
Astronomical unit, 27
Atlas (Saturn XV), 410, 709, 711
Atmospheric loading, 244, 248-249
Atomic time, 39, 84
Aurora, 493
Axis of figure, 14, 109, 110
Azimuth, 14
741
742
INDEX
B
Barycentric coordinates, 11
Barycentric Coordinate Time (TCB), 3, 45
Barycentric Dynamical Time (TDB), 2, 42
Belinda (Uranus XIV), 412, 709, 711
Besselian elements (see Eclipses)
Besselian solar year, 8
Bianca (Uranus VIII), 412, 709, 71 1
Brilliancy, 479
Pluto, 406
Saturn's rings, 404
c
Calendar, 575
accuracy, 578
Chinese, 577, 594, 599
Gregorian, 577, 580, 583, 604
Hebrew, 577, 584, 588
Indian, 591,594, 605
Islamic, 577, 589, 591, 604
Jewish (see Hebrew)
Julian, 600, 605
lunar, 577
lunisolar, 577
Moslem, (see Islamic)
reform, 578
solar, 577
Callisto (Jupiter IV), 708, 710
conjunctions, 348
orbital elements, 345
phenomena, 348
physical and photometric data, 409
Calypso (Saturn XIV), 410
state vector, 353, 708,710
Carme (Jupiter XI), 708, 710
Cartographic coordinates, 383-388
Galilean satellites, 408^109
longitude systems, 403
System HI, 403
planetocentric coordinates, 385-388
latitude, 387, 395
longitude, 387
range, 387
planetographic coordinates, 385-388
height, 387
latitude, 387, 395
longitude, 387, 395
reference surface, 385
prime meridian, 384
Mars, 403
Mercury, 402
Pluto, 406
Sun, 398
Saturn's satellites, 411
selenographic, 399^100
colongitude, 399
Catalog drift, 293
Catalogs
compiled, 505
differential, 506
fundamental, 506
observational, 505, 508
absolute, 507
differential, 507
fundamental, 507
reference, 507
Celestial coordinates, 20
Celestial ephemeris pole, 109, 110, 139
Celestial reference system, 199
Celestial sphere, 10
Ceres, 697
Cesium-beam standards, 60
Chandler wobble, 257
Charon (Pluto I), 709,711
orbital elements, 377
physical and photometric data, 414
Chebyshev polynomials, 305-307
Chronological cycles and eras, 579
Christian, 579
Hebrew, 584
Indian, 591
Islamic, 589
Circulation, 33 1
Clock performance, 58-61
Comets, 310-315
Commensurability, 330
Configurations, 476-482
Conjunctions, 33, 479
Connected-Element Radio Interferometry, 258
INDEX
743
Constants, 693
astronomical, 696
fundamental, 693
gravitational, 696
mathematical, 714
Constellations, 713
Conventional International Origin (CIO), 15,
53, 110, 139, 223
Conventional terrestrial reference system, 200,
223-224
Coordinated Universal Time (UTC), 6, 53, 76
Coordinates, astronomical, 202-207
barycentric, 11, 1 64
conversion from B1950.0 to J2000.0, 167,
184
geocentric, 202-207
geodetic, 202-207, 217-218
grid, 209-210
local, 207-216
planetocentric, 31, 37, 385-388
planetographic, 37, 385-388
reductions between apparent and mean
places, 145, 152
reference frame, 8
selenographic, 399
state plane, 214-216
topocentric, 10, 160-165
transformations, 555-559
Cordelia (Uranus VI), 412, 709, 7 1 1
Cressida (Uranus IX), 412, 709, 71 1
D
Data centers, 509
Centre de Donnees Stellaires, 509
Goddard Space Flight Center, 509, 513
Datums, 16, 216-223
Day corrections, 292
Day numbers, Besselian, 157, 169
Greenwich sidereal, 49, 75
independent, 157-158
Julian, 8, 55, 580, 600, 604
second-order, 158-160
Defect of illumination, 397
Deflection of light, 24
Deflection of the vertical, 204
Deimos (Mars II), 385, 708, 710
orbital elements, 343
physical and photometric data, 407
reference surface, 385
Delta T, 54, 81,83, 265-269
Desdemona (Uranus X), 412, 709, 71 1
Despina (Neptune V), 709, 711
Diameter
apparent polar, 397
asteroids, radiometric method, 414—417
equatorial, 397
polar, 396
Differential aberration, 134
Differential nutation, 120
Differential precession, 108
Dione (Saturn IV), 708, 710
orbital elements, 356
physical and photometric data, 410
Dip, 488
Distance, apparent, 392, 395, 396
Doppler method, 62
Doppler satellite tracking, 257-258
Double stars, 515-516
Dynamical time, 39, 41 , 54, 63
E
Earth, aphelion, 477, 479
figure, 200, 700
gravity field, 224-233
mass, 697
orbital elements, 316, 700, 704
orientation, 199
perihelion, 477, 479
physical and photometric data, 706
rotation, 25 1
rotation elements, 701, 705
rotation measurements, 61-63, 76
tides, 233-238, 283-284
Easter, 581, 583
Eccentricity, 27, 315-316, 704
744
INDEX
Eclipses, 34, 421
Bessel's method, 434
center of figure offset, 424
Circulars, 423
circumstances, 42 1
elements, 421
elements, orbitai, 3 1 5-3 1 6
ellipsoid, 200, 202, 206, 218-220, 223
ephemerides, 424-425
lunar, 35
Besselian elements, 468^169
contacts, 467, 469
magnitude, 470
obscuration, 469
occurrence, conditions for, 426-431
partial, 430, 467, 470
penumbra 1. 430, 467, 470
phases, 470
position angle, 470
total, 431, 467, 470
visibility, 470-47 1
penumbra, 428, 438
published data, 422-424
satellites of Jupiter, 348
solar, 35
annular, 433
annular-total, 433
Besselian elements (see elements)
central line, 455—456
circumstances, general, 421 , 446, 450,
459-461
circumstances, local, 461^167
contacts, 459-460, 462
curves, 450, 453-^58
duration, 455, 458
elements, 450
Besselian, 435-441
Besselian, auxiliary, 445, 449
elevation, correction for, 467
equations of condition, 446-450
limb corrections, lunar, 425
magnitude, 458, 463^164
maps, 423, 462
obscuration, 463, 465^466
observer's coordinates, 441-446
occurrence, conditions for, 426-428.
432-434
partial, 431
position angle, 463
total, 433
time scale (Delta T), 425, 467
umbra, 429, 438
Ecliptic, 11. 114
Elara (Jupiter VII J, 708, 710
state vector, 353
Elements, Besselian (see Eclipses and
Occupations)
Elements, orbital, 315-316, 704
Ellipsoid, 200, 202, 206, 218-220, 223
Elongations, 33
Enceladus (Saturn II), 708, 710
orbital elements, 356
physical and photometric data, 410
Energy conversion factors, 715
Ephemerides, fundamental, 279
initial conditions, 304
Ephemeris meridian, 54-55
Ephemeris time, 2-3, 42, 76, 79, 81-82
Ephemeris transit, major planets, 309
Epimetheus (Saturn XI), 410, 709, 71 I
Epochs, 27, 698
Equation of center, 327
Equations of motion, 26
Equation of the equinoxes, 4, 48, 53, 1 16
Equation of time, 5, 55, 74, 485
Equator, 1 1. 114,386,393
Equinox, catalog, 48-49, 54, 97, 293
dynamical, 1 1, 55, 97, 288-289
Equinox correction, 167-168
Equinox of date, 75
dates of, 477
Errors. 566-568
Euler angles, 284-285
Europa (Jupiter JJ). 708, 710
conjunctions, 348
orbital elements, 345
phenomena, 348
physical and photometric data, 409
F
Figure effects, 282
First point of Aries {see Equinox), 12
FK5, 167
INDEX
745
Flattening, major planets and Moon, 398-407
apparent, 390
Mercury, 402
Force model evaluation, 287
Frequencies, standard, 58
Fundamental ephemerides, 279-281
G
Galatea (Neptune VI), 709, 71 1
Galaxies, bright, 51 1
Second Reference Catalogue of Bright
Galaxies, 511,528
Third Reference Catalogue of Bright
Galaxies, 530
X-ray sources, 533
Galilean satellites, 345
Ganymede (Jupiter III), 708, 710
conjunctions, 348
orbital elements, 345
phenomena, 348
physical and photometric data, 409
Gaussian gravitational constant, 27
Gegenschein, 493
Geocentric coordinates, 10, 126, 202-207
Geocentric Coordinate Time (TCG), 3, 45
Geocentric planetary phenomena, 478
Geodetic coordinates, 16, 126, 202-207,
217-218
Geodetic datums, 702
Geodetic reference spheroids. 701
Geographic coordinates, 16
Geoid, 17
undulation, 202
Geopotential field, 227-233
Geopotenttal surface, 201
Global Positioning System (GPS), 64, 68,
259-261
Globular star clusters, 511, 526
Gravitational model. 280
Greek alphabet, 7 1 7
Greenwich civil time, 76
Greenwich mean astronomical time, 76
Greenwich mean time, 7, 76
Greenwich meridian, 384, 386. 393
Greenwich sidereal angle, 21
Greenwich sidereal date, 49, 75
Greenwich sidereal time, 4, 49-53, 1 16
Greenwich transit, 54
Gregorian calendar, 7
Gyroscopes, 261-262
H
Helene (Saturn XII), 410, 709, 71 1
Heliocentric coordinates, 1 1
Heliocentric planetary phenomena, 481-482
Himalia (Jupiter VI), 708, 710
state vector, 353
Horizon system, 208-209
Horizontal parallax
Moon, 400
Sun, 398
Hubble sequence, 528
Hydrogen masers, 60
Hyperion (Saturn VII), 385, 41 1, 708, 710
differential coordinates, 367
orbital elements, 360
physical and photometric data, 385, 4 1 1
I
Iapetus (Saturn VIII), 411, 708, 710
differential coordinates, 367
orbital elements, 360
physical and photometric data, 41 1
Illuminated disk, 388-397
Illumination, 490-492
planets, 401-407
Illumination, defective, 397
Improved Lunar Ephemeris, 81
Inclination, 27
Initial conditions, 304
International Atomic Time (TAI), 2, 40
International Earth Rotation Service (IERS),
86, 111, 200, 223-224, 262-265
Interpolation, 546-548
Invariable plane, 1 1
Inverse interpolation, 545, 548
Io (Jupiter I), 708,710
conjunctions, 348
orbital elements, 345, 708
phenomena, 348
physical and photometric data, 409, 710
746
INDEX
J
janus (Saturn X), 410, 708, 710
Julian calendar, 7
Julian date, 8, 55, 600
modified, 8
Julian day numbers, 8, 55, 580, 600, 603
Julian period, 580
Julian year, 8
Juliet (Uranus XI), 412, 709, 711
Juno, 697
Jupiter
magnitude, 404
mass, 697
orbital elements, 316, 704
phase, 404
physical and photometric data, 706
physical ephemeris, 403-404
rotation elements, 404, 705
satellites, 345, 409, 708, 710
Jupiter's satellites, 708, 710
I-IV, 345, 409
V, 348, 409
VI-XHI, 352, 409
XIV, 351
K
Kepler's law, 26
L
Laplacian plane, 327
Larissa (Neptune VII), 709, 71 1
Laser ranging, 254-255
Latitude (see also Coordinates)
planetocentric, 387, 395
planetographic, 387, 395
Laws of motion, 24
Leap second, 7, 53
Least squares, 300-301, 568-573
Leda (Jupiter XIII), 708, 710
state vector, 353
Length of day, 19, 265-269
Librations, Moon, 398^100
geocentric optical, 398
physical rotational, 398
topocentric optical, 399-400
Librations, satellite, 33 1
Light, deflection of, 135-139, 166
velocity of, 693
Light-time, 127, 148
correction for, 23, 393
Local apparent solar time, 4
Local coordinates, 207-216
Local hour angle, 14, 55
Local mean time, 5, 76
Local meridian, 1 1, 55
Local place, 165
Local sidereal time, 4, 49, 52-53
Longitude {see also Coordinates)
planetocentric, 387
planetographic, 387, 395
Longitude of node, 27
LORAN-C, 66
Love numbers, 31, 233, 250
Luminosity, Iapetus, 41 1
Luminosity class, bright galaxies
planetocentric, 387
planetographic, 387, 395
Lunar eclipses (see Eclipses, lunar)
Lunar ephemeris, 317
Lunar Laser-Ranging, 62, 299-300
Lunar librations, 284-286, 398^400
geocentric optical, 398
physical rotational, 398
topocentric optical, 399-400
Lunar secular acceleration, 288
Lysithea (Jupiter X), 708, 710
state vector, 353
M
Magnitude
asteroids, 414-416
infrared, 415
visual, 414
major planets (see individual planets)
at 1 AU and 0° phase, 388
mean opposition, 389
minor planets, 4144-16
infrared, 415
visual, 414
Saturn's rings, 404
INDEX
Mars
defective illumination, 403
magnitude, 403
mass, 697
orbital elements, 316, 704
physical and photometric data, 706
physical ephemeris, 403
rotation elements, 403, 705
satellites, 342, 403, 708, 710
Masses, 697
Mathematical model, 281
Matrices, 552-559
Matrix methods, 1 80
Maximum defect of illumination, 397
Mean sea level, 201-202
Mean sidereal time, 4, 48, 75
Mean solar time, 3, 74-75, 79
Mercury
brilliancy, 402
elongations, 479
magnitude, 402
mass, 697
orbital elements, 316, 704
physical and photometric data, 402, 706
physical ephemeris, 401^103
rotation elements, 402, 705
transit, 47 1
Mercury ion frequency standard, 61
Meridian passage (see Transit, meridian, 33
Metis (Jupiter XVI), 708, 710
Mimas (Saturn I), 708, 710
orbital elements, 356
physical and photometric data, 410
Minor Planet Center, 310
Minor planets, 310-315
geocentric phenomena, 414-417
magnitudes, 3 1 1
Miranda (Uranus V), 709, 71 1
orbital elements, 368
physical and photometric data, 412
MK spectral classification system, 520
Modified Julian date, 56
Month, 577
lengths, 698
synodic, 486, 576
Moon
defective illumination, 399
eclipses (see Eclipses, lunar)
ephemeris, basis of, 317
fraction illuminated, 399
horizontal parallax, 400
librations, 398-400
optical, 398
physical, 398
topocentric, 399-400
limb corrections, 425
mass, 696
mean longitude, 115
midpoint of bright limb, 399
node, 1 1 5
orbital elements, 701, 708
perigee, 115
phases, 478^179
physical and photometric data, 398^101,
701,706,710
physical ephemeris, 398—401
principal moment of inertia, 399
rise and set, 485-486
secular acceleration, 77, 84
selenographic coordinates, 399-400
semidiameter, 401
topocentric parallax, 400
Moonlight, 493
Moonrise and moonset, 32, 485^-86
N
Naiad (Neptune III), 709, 71 1
Nautical twilight, 33, 483, 492
Neptune
magnitude, 406
mass, 697
orbital elements, 316, 704
physical and photometric data, 405-406,
706
physical ephemeris, 405-406
rotation elements, 705
satellites, 373, 4 1 3^1 1 4, 709, 7 1 1
Nereid (Neptune II), 709, 71 1
differential coordinates, 376
orbital elements, 375
Neweomb's Tables, 317
Nonrigid-body effects, 286
748
INDEX
North pole, 384-385, 392-393
apparent, 391
Sun, 398
Numerical differentiation, 560-562
Numerical integration, 27, 286, 562-566
Nutation, 12,99, 109, 111, 116-121, 151
corrections, 116-119
differential, 120
IAU theory, 111-113
in longitude and obliquity, 112-1 15
planetary terms, 117-119
o
Oberon (Uranus IV), 709, 7 1 1
oblateness, 14
orbital elements, 369
physical and photometric data, 412
Obliquity of the ecliptic, 12, 171, 289
Observational catalog, 179
Occupations, 34, 494-498
Occupations, lunar, 494
elements (Besselian), 494
planets, 498
Ocean loading, 244-248
Ocean tide model, 240-244
Open star clusters, 51 1, 522
Lund Catalog, 511,522
Trumpler classification system, 524
Ophelia (Uranus VII), 412, 709, 711
Opposition, 33
Orbital elements, 27
P
Pallas, 697
Pandora (Saturn XVII), 410, 709, 7 1 1
Parallax, 21, 121, 123
Parallax, annual, 22, 125
diurnal (geocentric), 22, 123-124, 160
equatorial horizontal, Moon, 400
horizontal, Moon, 400
horizontal, Sun, 398
in altitude, 125
solar, 22
topocentric, Moon, 400
Parameterized post-Newtonian equation, 281
Pasiphae (Jupiter VIII), 708, 710
state vector, 353
Perihelion, 477
Perturbations, 28
Phase, 396
angle, 388, 396
coefficient, 388
Galilean satellites, 408-409
Mercury, 402
Venus, 403
corrections, 291-292
fraction of Moon illuminated, 399
Moon, 399, 478
Phases and magnitudes, 388-397
inlrinsic properties, 388
Phobos (Mars I), 385, 708, 710
orbital elements, 343
physical and photometric data, 385, 407
Phoebe (Saturn IX), 708, 710
differential coordinates, 368
orbital elements, 367
physical and photometric data, 410
Photometric standards, stars, 51 1, 516
UVBRI, 51 1,517
uvby and H/3, 511,517
Physical ephemerides, 36, 383^419
asteroids, 414-417
Moon, 398^101
planets, 401-407
Jupiter, 403-104
Mars, 403
Mercury, 401^102
Neptune, 405^106
Pluto, 406
Saturn, 404-405
Uranus, 405
Venus, 402^03
references, 417^119
satellites, 407-11 4
Galilean, 408^109
Jupiter, 408-109
Mars, 407-408
Neptune, 413^114
Pluto, 414
Saturn, 408,410-411
Uranus, 411-413
Sun, 397-398
Planetocentric coordinates, 31, 37, 385-388
Planetocentric orbital longitude, Sun, 401
INDEX
749
Planetographic coordinates, 37, 385-388
Planets
aberration, 150
apparent positions, 145-152
astrometric positions, 166, 308
brightness
Pluto, 406
Saturn's rings, 404
brilliancy, 479
configurations, 476-482
deflection of light, 149
elements, 315-316, 704
elongations, 479
geocentric phenomena, 478^181
heliocentric phenomena, 481^-82
masses, 697
orbital elements, 315-316, 704
physical and photometric data, 401—407,
706
precession, 150
rings, 325, 712
rotation elements, 705
satellites, 325, 407-414, 708-71 1
semidiameters, 401-407
stationary, 479
synodic periods, 480-481
visibility, 480
Plate motions, 249-250
Pluto
mass, 697
orbital elements, 316, 704
physical and photometric data, 406, 706
physical ephemeris, 406
rotation elements, 406, 705
satellite, 377,414, 709,711
Point mass interactions, 281
Polaris, 498-502
Polar motion, 14, 18, 109, 111, 139-140,
238-240, 269^272
Polynomials, 548
Portable clocks, 65
Portia (Uranus XII), 412, 709, 71 1
Positions
angle, 392, 395
bright limb, Moon, 399
greatest defect of illumination, 397
apparent, 23, 145-160, 165
astrometric, 23, 166, 308, 311-313
geocentric, 10, 126, 202-207
heliocentric, 1 1
Precession, 105, 150
differential, 108
drift, 293
general, 99
geodesic, 100
IAU 1976, 173
luni-solar, 12, 99
Newcomb's, 106, 173
planetary, 13, 99
reduction, day number formulas, 108
rotation matrix, 103
Prime meridian, 384, 386, 393
Mars, 403
Mercury, 402
Pluto, 406
Sun, 398
Prime vertical, 204
Prometheus (Saturn XVI), 410, 709, 711
Proper motion, 22, 121, 171. 174, 184
Proteus (Neptune VIII). 709, 71 1
Puck (Uranus XV), 412, 709, 71 1
Pulsars, 511,521
Q
Quartz-crystal oscillators, 60
Quasars, 511, 530
Quasi-stellar objects, 530
active galactic nuclei, 531
catalogs, 530-531
Radar ranging, 294-297
Radial velocity, 22, 1 2 1 . 5 1 1 . 5 1 8
Radio interferometry, 62
Radiometric method of diameter estimation,
414^117
geometric albedo, 415
infrared magnitudes, 415
phase integral, 416
Radio sources
astrometric calibrators, 51 1, 531
flux calibrators, 511,532
Radius of the planets, 387
apparent polar, 390, 396
apparent subsolar, 396
Reference frame, inertial, 96
Reference frames, 223, 280
750
INDEX
Reference systems, 95-99. 199
Refraction, 24, 140-145
rising and setting phenomena. 488
Saastamoinen, 143
Relativity, 40, 70, 96, 149, 295
Rhea (Saturn V),410, 708,710
orbital elements, 360
physical and photometric data, 410
Rings of Satum, brilliance, 404
Rising and setting phenomena, 482^190
Moon, 485^*86
Sun, 483-485
Rosalind (Uranus XIII), 412. 709. 711
Rotational elements, 383-388
deformation, 238-240
direct, 385
retrograde, 385
Sun, 397
Rotation of axes, 552-553
Rotation of the Earth, 25 1
Rubidium clocks, 61
s
Saros, 35
Satellite Almanac, 325
Satellites
apparent distance and position angle, 341
apparent orbits, 336
authorities, 378
differential coordinates, 341
elongations, 341
invariable plane, 327
Laplacian plane, 327
orbital data, 708
planetocentric coordinates, 333
rectangular coordinates, 341
Saturn
mass, 697
orbital elements, 3 1 6, 704
physical and photometric data, 404-405,
706
physical ephemeris, 404-405
rings, 354, 404, 712
rotation elements, 404, 705
satellites, 356, 408, 410-41 1, 708-71 1
semidiameter, 404
Schwiderski model, 242-243
Seasons, 477, 479
Second, SI, 40, 84
Selenocentric coordinates, 1 1
Selenographic coordinates, 399
Semidiameter
Moon, 401
Sun, 398
Sidereal day, 48
Sidereal hour angle, 1 1
Sidereal time, 3^*. 39, 48, 55
apparent, 4, 48
local, 4, 49
mean, 4, 48
relation to universal time, 50-53
Sinope (Jupiter IX), 708, 710
Shadow lengths, 489
State vector, 353
Solar corona, 295
Solar eclipse (see Eclipses, solar)
Solar parallax, 398
Solid Earth tides, 233-238
Solstices, times of, 477
Spacecraft data, 297-299
Space motion, 121
Space-Time coordinates, 45
Spectral classification, 519
MK spectral standards, 51 1-520
Morgan-Keenan classification system
(MK), 520
Spherical astronomy, 543, 549-552
Spherical coordinates, 9, 553
Spherical triangles, 542
Spherical trigonometry, 554
Star clusters
globular, 511,526
open, 511,522
Starlight, 493
Stars
apparent places, 23, 152-160, 510
bright, 51 1-512
double, 515-516
foreshortening, 122
mean places, 51 1
parallax, 121, 123, 398
photometric standards, 511,516
proper motion, 121
INDEX
751
Stars (continued)
radial velocity standards, 51 1, 518
secular variations, 121
spectral classification, 519
MK spectral standards, 511, 520
Morgan-Keenan system (MK), 520
stellar positions, 184
stratosphere, 141
variable, 5 1 1
Stationary points, 479
Statistics, 566-568
Stellar data, 509
Stellar positions, 184
Sub-Earth point. 391-395
Subsolar point, 392-395
Subtabulation, 546
Sun, 700
eclipses (see Eclipses, solar)
ephemeris, basis of, 317
low precision ephemeris, 484
mass, 697
physical ephemeris, 397-398
rise and set, 483-485
rotation elements, 705
Sundials, 73
Sunrise and Sunset, 32, 483-485
Surface brightness, 390
Surface brightness, planets, 390
Synodic periods, 480-48 1
System III longitudes, Jupiter, 403
T
TAI (International Atomic Time), 2, 39, 84
Telesto (Saturn XIII), 410
Television transmissions, 67
Terminator, geometric, 397
Terrestrial coordinates, 13, 20, 126
Terrestrial Dynamical Time (TDT), 2, 42, 54
Terrestrial reference system, 199, 200,
217-218
Terrestrial Time (TT), 3, 45
Tethys (Satum III)
orbital elements, 356, 708
physical and photometric data, 410, 710
Thalassa (Neptune IV), 709. 7 1 1
Thebe (Jupiter XIV), 708, 710
orbital elements, 351
Tidal friction, 19
Time (see specific names)
equation of, 5, 55, 74, 485
signals, 65
transfer, 64
zones, 56-57
Titan (Saturn VI)
orbital elements, 360, 708
physical and photometric data, 4 1 0, 7 1 0
Titania (Uranus III)
orbital elements. 369, 709
physical and photometric data, 412. 71 1
topopause. 141
troposphere, 295
Topocentric coordinates, 10, 160-165
Topocentric parallax, Moon, 400
Transformations to FK5, 167-187
matrix method, 180
Transit across Sun's disk, 47 1
planets, 471
Transit circle data, 290-294
Transit, ephemeris, 309
Transits, 34
Trigonometric functions, 544
Triton (Neptune I)
orbital elements, 373
physical and photometric data, 413
troposphere, 141
Tropical year, length, 80
Twilight, 33, 482, 492
astronomical, 33, 483, 492
civil, 33, 483, 492
nautical, 33,483,492
u
UBVRI photometric standards, 511,517
Umbriel (Uranus II)
orbital elements, 369, 709
physical and photometric data, 412, 71 1
Unit, astronomical, 27
Units
International System, 718-719
length, 716
mass, 7 16
speed, 716
Universal Polar Stereographic, 210-214
752
Universal Time (UT), 3, 39, 50-54, 76-77
relation to sidereal time, 5 1-52
tidal effects, 250-254
Universal Time Coordinated (UTC), 6, 53, 76,
85-86
Universal Transverse Mercator, 210-214
Uranus
mass, 697
orbital elements, 316, 368, 704
physical and photometric data. 706
physical ephemeris, 405
rings, 368. 712
rotation elements, 405, 705
satellites, 41 1—413, 709, 711
semidiameter, 405
uvby and H/.3 photometric standards, 511, 517
X-ray sources, 511, 533
catalogs, 533
Year, 576, 698
lengths, 576, 698
lunar, 577
sidereal, 698
tropical, 80, 576
Zodiacal light, 493
Zonal tide terms, 252-253
V
Variable stars, 511,513
American Association of Variable Star
Observers (AAVSO), 513-514
General Catalog of Variable Stars (GC VSJ,
513
Name Lists of Variable Stars, 5 1 4
Venus
brilliancy, 402, 479
magnitude, 402
mass, 697
orbital elements, 316, 704
physical and photometric data, 402^103,
706
physical ephemeris, 402^-03
rotation elements, 402, 705
semidiameter, 402
transit of, 471
Very Long Baseline Interferometry (VLBI),
69, 255-256
Vesta, 697
Virtual place, 165
w
Week, 577, 603
WWV, 86
»EX